AUTOMATION WORKFLOW

From daily news and career tips to monthly insights on AI, sustainability, software, and more—pick what matters and get it in your inbox. Discover the engineering revolution transforming modern defense with Strength, Stealth, Speed: The Very Fast Future of Advanced Defense Access expert insights, exclusive content, and a deeper dive into engineering and innovation all with fewer ads or a completely ad-free experience. All Rights Reserved, IE Media, Inc. Follow Us On Future of Defense Access expert insights, exclusive content, and a deeper dive into engineering and innovation all with fewer ads or a completely ad-free experience. All Rights Reserved, IE Media, Inc. The Seoul fashion show explored how humans and robots may coexist in everyday life. Humanoid robots wearing designer outfits walked alongside human models at a fashion show in Seoul this week, offering a glimpse into how South Korea’s technology industry imagines a future where robots are not just tools, but participants in everyday cultural life. The event, called the “Mach33: Physical AI Fashion Show,” was hosted by South Korean entertainment technology company Galaxy Corporation and featured robots and humans walking the runway together in coordinated outfits. Videos and images from the show depicted humanoid robots strutting down the catwalk, posing beside models, and performing synchronized choreography. According to Reuters, the event was designed around the idea of humans and robots coexisting in daily life, with matching outfits intended to imagine how future interactions between people and physical AI systems might look. The fashion show took place at Galaxy Robot Park in Seoul, a recently opened robot-themed entertainment complex that combines robotics, artificial intelligence, K-pop culture, and interactive attractions. Humanoid robots have traditionally been demonstrated in controlled industrial settings, research laboratories, or technology exhibitions focused on engineering capabilities. The Seoul event reflected a growing shift toward presenting robots in social, cultural, and entertainment environments rather than purely technical ones. According to Reuters and local South Korean media reports, the runway presentation featured robots dressed in designer clothing and paired with human models, with organizers describing the concept as a “physical AI” showcase. Galaxy Corporation has increasingly positioned itself as an “enter-tech” company that combines entertainment and advanced technology. The firm is also known for managing major Korean entertainment figures, including K-pop star G-Dragon. The fashion show was part of a broader effort by the company to expand robot-centered entertainment experiences. Reports indicate Galaxy also plans robot concerts, interactive performances, and additional AI-focused cultural events. The runway event arrives as South Korea continues to strengthen its position as one of the world’s most robot-intensive economies. South Korea has one of the highest robot densities globally, with more than 1,000 industrial robots for every 10,000 workers. The country has invested heavily in automation, advanced manufacturing, artificial intelligence, and humanoid robotics. In recent years, South Korea has also launched major initiatives to accelerate domestic humanoid robot development, including collaborations among technology firms, universities, and government-backed research programs. At the same time, robotics companies worldwide are increasingly attempting to move humanoid machines beyond factories and warehouses into environments where they interact more directly with people. That transition remains difficult. While modern humanoid robots have become significantly better at walking, balancing, dancing, and performing choreographed movements, researchers continue to face challenges involving dexterity, autonomy, perception, and natural human-robot interaction. Still, events like the Seoul runway show demonstrate how robotics is increasingly being presented not just as an industrial technology, but as part of broader discussions about culture, design, entertainment, and daily life. Whether robot fashion shows become a lasting trend or remain a technological novelty, the sight of humanoid machines sharing the catwalk with human models shows how quickly robotics is moving into spaces once considered uniquely human. Kaif Shaikh is a journalist and writer passionate about turning complex information into clear, impactful stories. His writing covers technology, sustainability, geopolitics, and occasionally fiction. A graduate in Journalism and Mass Communication, his work has appeared in the Times of India and beyond. After a near-fatal experience, Kaif began seeing both stories and silences differently. Outside work, he juggles far too many projects and passions, but always makes time to read, reflect, and hold onto the thread of wonder. Premium Follow

AI robotics company Figure posted several videos on X throughout May showcasing its robots performing basic tasks, including cleaning a room and sorting packages. Modern artificial intelligence-powered robots are impressive in their capabilities, but are still years away from replacing humans as they can’t yet adapt to changing conditions, researchers say. Last month, AI robotics company Figure showcased its humanoid robots performing basic tasks, such as cleaning a room, but a series of robots working for nine days straight sorting packages sparked conversation about how soon robots could replace jobs. Oliver Obst, an associate professor of robotics at the Australia based University of New South Wales, told Cointelegraph that repetitive jobs such as physical work in structured environments are currently most at risk of being replaced by robots, while administrative and document-processing tasks could be replaced by AI. There has been growing concern that AI and robots will replace people in jobs as technology advances. A report in May from workforce consulting firm Challenger, Gray and Christmas found that US companies have laid off an estimated 49,135 people in 2026 due to AI. A group of Figure’s robots worked for nine days straight sorting packages. Source: Figure However, Obst said that humanoid robots are unlikely to see a mass rollout soon because they don’t appear to be more efficient or less error-prone than current robotic manufacturing methods. “Even in relatively structured settings, they still face problems with reliability, speed, safety, cost, and recovery from unexpected situations,” he said. “The harder the environment is to control, the harder the robotics problem becomes. Most human jobs involve more variation and more judgment than the package-sorting demonstration.” In another video in May, a human worker managed to sort more packages compared to a team of Figure’s robots, which swapped out when needing a recharge. Figure CEO Brett Adock said it would be the last time “a human will ever win.” Source: Brett Adock Markus Levin, co-founder of decentralized data network XYO, said AI models and automation software can perform repetitive tasks with far greater consistency and endurance than humans; however, robots still require charging, maintenance and supervision. A report in September from the International Federation of Robotics found that global demand for factory robots has doubled over the last decade, with warehouses and logistics among the fastest-growing areas of adoption. “I believe broad human replacement is still likely years away,” Levin added, “Reliability, safety, regulation, infrastructure costs, and trust remain major barriers to full-scale deployment across society. The challenge is no longer simply making machines capable of acting but ensuring they can operate safely and reliably as they take on greater autonomy.” Dr Francisco Cruz Naranjo, a senior lecturer at the University of New South Wales with a PhD in robotics, said the efficiency of robots compared to people depends heavily on the activity and the environment. Related: ‘Developed ecosystem’ based on crypto has sprung up for AI agents: Report “Robots are much better at repetitive tasks without the need for constant pauses, as showcased in the Figure livestream. However, in highly dynamic environments, robots still struggle to quickly adapt to changing conditions,” he said. Naranjo said repetitive jobs performed in a less static setting are at risk of being replaced by robots, but it will depend on how quickly research advances and how quickly society adapts in areas like making spaces robot-friendly, which is likely years away. Naranjo and Obst said that a mass rollout of robots in the workforce could be of some benefit, such as improving work-life balance, increasing the workforce in areas with shortages, and addressing dangerous environments that are too risky for humans. “The social question is harder. If robots make dangerous work cheaper in human terms, that can be good. But it can also have unintended consequences. For example, keeping humans out of harm’s way in military operations may save lives, but it could also lower the perceived cost of conflict,” Obst said. Magazine: Korea’s first memecoin rug-pull case, China’s crypto rules review: Asia Express More on the subject Cointelegraph is committed to providing independent, high-quality journalism across the crypto, blockchain, AI, and fintech industries. All news, reviews, and analyses are produced with full journalistic independence and integrity. For more details on our standards and processes, please read our Editorial Policy.

Authored by Stephen Katte via Cointelegraph, AI robotics company Figure posted several videos on X throughout May showcasing its robots performing basic tasks, including cleaning a room and sorting packages. Modern artificial intelligence-powered robots are impressive in their capabilities, but are still years away from replacing humans as they can't yet adapt to changing conditions, researchers say. Last month, AI robotics company Figure showcased its humanoid robots performing basic tasks, such as cleaning a room, but a series of robots working for nine days straight sorting packages sparked conversation about how soon robots could replace jobs. Welcome to Day 9 of our humanoid livestream: 191 consecutive hours and 238,000 packages. Oliver Obst, an associate professor of robotics at the Australia based University of New South Wales, told Cointelegraph that repetitive jobs such as physical work in structured environments are currently most at risk of being replaced by robots, while administrative and document-processing tasks could be replaced by AI. There has been growing concern that AI and robots will replace people in jobs as technology advances. A report in May from workforce consulting firm Challenger, Gray and Christmas found that US companies have laid off an estimated 49,135 people in 2026 due to AI. However, Obst said that humanoid robots are unlikely to see a mass rollout soon because they don't appear to be more efficient or less error-prone than current robotic manufacturing methods. "Even in relatively structured settings, they still face problems with reliability, speed, safety, cost, and recovery from unexpected situations," he said. "The harder the environment is to control, the harder the robotics problem becomes. Most human jobs involve more variation and more judgment than the package-sorting demonstration." "I would not say we are at the point of mass replacement by humanoid robots. We are much closer to the selective automation of some tasks. AI software is moving faster and is already affecting some forms of information work, but physical robots still have a much harder problem to solve." In another video in May, a human worker managed to sort more packages compared to a team of Figure's robots, which swapped out when needing a recharge. Figure CEO Brett Adock said it would be the last time "a human will ever win." Congrats to Aime!! He said his left forearm is basically broken. Final scores: F.03: 12,732 packages (2.83 seconds/package) - Aime: 12,924 packages (2.79 seconds/package). This is the last time a human will ever win. Markus Levin, co-founder of decentralized data network XYO, said AI models and automation software can perform repetitive tasks with far greater consistency and endurance than humans; however, robots still require charging, maintenance and supervision. A report in September from the International Federation of Robotics found that global demand for factory robots has doubled over the last decade, with warehouses and logistics among the fastest-growing areas of adoption. "I believe broad human replacement is still likely years away," Levin added, "Reliability, safety, regulation, infrastructure costs, and trust remain major barriers to full-scale deployment across society. The challenge is no longer simply making machines capable of acting but ensuring they can operate safely and reliably as they take on greater autonomy." Dr Francisco Cruz Naranjo, a senior lecturer at the University of New South Wales with a PhD in robotics, said the efficiency of robots compared to people depends heavily on the activity and the environment. "Robots are much better at repetitive tasks without the need for constant pauses, as showcased in the Figure livestream. However, in highly dynamic environments, robots still struggle to quickly adapt to changing conditions," he said. "Humans, in this case, are much better. This is precisely why robots at the moment are highly efficient in controlled environments, such as factories, but they have not yet succeeded widely in home settings." Naranjo said repetitive jobs performed in a less static setting are at risk of being replaced by robots, but it will depend on how quickly research advances and how quickly society adapts in areas like making spaces robot-friendly, which is likely years away. Naranjo and Obst said that a mass rollout of robots in the workforce could be of some benefit, such as improving work-life balance, increasing the workforce in areas with shortages, and addressing dangerous environments that are too risky for humans. "The social question is harder. If robots make dangerous work cheaper in human terms, that can be good. But it can also have unintended consequences. For example, keeping humans out of harm's way in military operations may save lives, but it could also lower the perceived cost of conflict," Obst said. "Hypothetically, if we became very successful at automating almost all work, then society would need to rethink economies that are currently built around individual wages and employment." Assistance and Requests: Contact Us Tips: tips@zerohedge.com General: info@zerohedge.com Legal: legal@zerohedge.com Advertising: Contact Us Abuse/Complaints: abuse@zerohedge.com Make sure to read our "How To [Read/Tip Off] Zero Hedge Without Attracting The Interest Of [Human Resources/The Treasury/Black Helicopters]" Guide It would be very wise of you to study our privacy policyand our (non)policy on conflicts / full disclosure.Here's our Cookie Policy. How to report offensive comments Notice on Racial Discrimination.

Geeky Gadgets The Latest Technology News 7:00 am May 30, 2026 By Julian Horsey Google’s decision to sell Boston Dynamics in 2017 underscored the tension between research-driven robotics and the demands of commercial viability. As Chromeborne explains, the sale stemmed from a mismatch between Boston Dynamics’ emphasis on experimental advancements, such as the humanoid robot Atlas and Google’s focus on creating products with clear and immediate market applications. This case illustrates the broader challenge of balancing long-term technological exploration with the pressures of short-term business goals. Explore how Boston Dynamics shifted its priorities under new ownership, including its move toward warehouse automation and logistics. Gain insight into the roles played by SoftBank and Hyundai in shaping the company’s trajectory. Understand the broader implications of integrating advanced robotics into industries that demand both innovation and profitability. TL;DR Key Takeaways : Boston Dynamics was established in 1992 by Marc Raibert, a visionary in the field of robotics. His ambition was to create machines capable of mimicking the agility, balance and movement of animals. From its inception, the company concentrated on dynamic locomotion, pushing the boundaries of what robots could achieve. Early projects were heavily research-driven, often funded by organizations like the Defense Advanced Research Projects Agency (DARPA). These initiatives prioritized technological breakthroughs over immediate commercial applications, solidifying Boston Dynamics’ reputation as a leader in innovation. The company’s early work laid the groundwork for its future success. By focusing on solving complex problems in robotic movement, Boston Dynamics developed technologies that would later influence the broader robotics industry. However, this emphasis on research over practicality also posed challenges, particularly when it came to finding real-world applications for their innovations. Boston Dynamics achieved several key breakthroughs that demonstrated the potential of advanced robotics. These milestones not only showcased the company’s technical expertise but also highlighted the challenges of translating innovation into practical use: While these innovations were new, their lack of immediate, practical applications hindered their commercial potential. This gap between technological achievement and market readiness became a recurring challenge for Boston Dynamics. Expand your understanding of Boston Dynamics with additional resources from our extensive library of articles. In 2013, Google acquired Boston Dynamics as part of its broader exploration into robotics and automation. At the time, Google was investing heavily in emerging technologies, aiming to position itself as a leader in the field. However, the partnership quickly revealed a fundamental misalignment of priorities. Google sought to develop robots that could address immediate industrial needs, focusing on market-ready solutions. In contrast, Boston Dynamics remained committed to long-term innovation and experimental prototypes. This divergence in goals created tension between the two companies. Boston Dynamics’ research-driven approach did not align with Google’s commercial ambitions, leading to friction and ultimately the decision to sell the company in 2017. The sale highlighted the challenges of integrating a research-focused organization into a commercially driven enterprise. When SoftBank acquired Boston Dynamics in 2017, the company began to pivot toward practical, real-world applications. This shift was evident in the development of quieter, electric-powered robots like Spot, which was designed to perform tasks in various industries. Spot’s capabilities included: Spot’s versatility and ability to navigate complex environments made it a valuable tool for industries such as construction, energy and manufacturing. Additionally, Boston Dynamics’ acquisition of Kinema Systems, a company specializing in robotic vision, enhanced its robots’ autonomy and adaptability, further aligning the company with market demands. In 2021, Hyundai acquired Boston Dynamics, marking another significant turning point in the company’s evolution. With Hyundai’s backing, Boston Dynamics intensified its focus on aligning its innovations with industrial needs. This partnership emphasized the development of robots for factory and warehouse automation, areas where robotics could deliver immediate value. Under Hyundai’s ownership, Boston Dynamics expanded the applications of its robots. Spot became a reliable tool for inspection tasks, while Atlas demonstrated potential for performing repetitive labor in controlled environments. This shift toward practical, real-world uses allowed Boston Dynamics to transition from a research-focused organization to a commercially viable enterprise. Hyundai’s support also provided the resources needed to scale production and refine the company’s technologies for broader adoption. Boston Dynamics faced numerous challenges throughout its journey, including: The company addressed these challenges by focusing on autonomy, environmental awareness and electric-powered designs. By adapting its technologies to meet market demands, Boston Dynamics successfully bridged the gap between innovation and practicality. This approach not only ensured the company’s survival but also solidified its position as a leader in the robotics industry. Today, Boston Dynamics is recognized as a pioneer in robotics, producing machines with clear, defined purposes. Spot has become a staple in industrial inspections, while Atlas continues to evolve as a platform for repetitive tasks in controlled environments. The company’s transformation from a research-driven organization to a commercially focused business underscores its ability to adapt and thrive in a competitive and rapidly evolving industry. As robotics technology continues to advance, Boston Dynamics remains at the forefront, shaping the future of automation and dynamic locomotion. Its journey serves as a powerful example of how innovation, when balanced with practicality, can drive progress and redefine what is possible in the field of robotics. Media Credit: Chromeborne Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Tinah F. Publié le 15 mai 2026 Mis à jour le 19 mai 2026 2 minutes de lecture Robotique Le robot Figure 03 de Figure AI vient de réaliser une démonstration qui fait parler dans le monde de la robotique. Il a effectué plus de 40 heures de tri de colis autonome sans pause ni assistance humaine. Ce qui est impressionnant dans cette histoire, ce n’est pas seulement qu’un robot humanoïde sache déplacer des colis sans tout faire tomber au bout de trois minutes. Le vrai sujet, c’est l’autonomie. Le robot Figure 03 aurait travaillé plus de 40 heures d’affilée sans interruption, sans assistance humaine et sans pause improvisée devant une machine à café inexistante. Figure AI veut donc prouver l’endurance de ses machines. Derrière cette démonstration, il y a surtout Helix-02, le nouveau réseau neuronal développé par Figure AI. C’est lui qui pilote les capacités des robots Figure 03 pendant ces longues sessions de travail. Le point mis en avant par l’entreprise n’est pas seulement la précision des mouvements. Figure insiste surtout sur la continuité du service. Les robots peuvent détecter certaines erreurs et reprendre automatiquement une tâche interrompue. Ils seraient aussi capables de gérer le remplacement de leurs batteries grâce à plusieurs unités fonctionnant en relais. ⚡️ INSIGHT: Figure says its robot crossed 30 hours of continuous autonomous work with no downtime. pic.twitter.com/4uAcPmoUYV Autrement dit, l’objectif n’est plus simplement de fabriquer un robot qui sait déplacer une boîte. Le vrai défi consiste maintenant à maintenir un système autonome pendant des dizaines d’heures dans un environnement industriel réel. Et c’est précisément là que les choses deviennent intéressantes. Parce qu’entre une vidéo virale sur X et une chaîne logistique qui tourne jour comme de nuit, il y a un gouffre technique et financier. Figure AI affirme également avoir expédié 350 robots depuis son usine BotQ de Sunnyvale, avec un rythme d’environ un robot produit par heure. Ces chiffres montrent surtout l’ambition industrielle de la société. Figure AI commence à se faire de la place dans le domaine de la robotique. Fondée seulement en 2022 par Brett Adcock, la société s’est rapidement imposée dans une course dominée par des géants comme Tesla ou Boston Dynamics. L’arme principale de Figure AI, c’est évidemment le robot Figure 03. Cet humanoïde de troisième génération représente une énorme évolution par rapport aux anciens prototypes de la marque. Beaucoup de robots humanoïdes restent encore limités à des démonstrations très contrôlées. De son côté, Figure 03 cherche surtout à prouver qu’il peut fonctionner dans des environnements réels. Et l’entreprise multiplie les démonstrations pour le montrer. Ces derniers mois, Figure AI a notamment diffusé une vidéo où l’on voit le robot ranger une chambre, déplacer des objets et organiser l’espace avec des gestes étonnamment fluides. 28 mai 2026 26 mai 2026 25 mai 2026 Rejoignez nos 100 000 passionnés et experts et recevez en avant-première les dernières tendances de l’intelligence artificielle🔥 Accueil > Robotique > Vous vous souvenez du robot Figure 03 ? Il travaille maintenant 40 heures d’affilée Rejoignez nos 100 000 passionnés et experts et recevez en avant-première les dernières tendances de l’intelligence artificielle🔥 Rejoins nos 100 000 passionnés et experts et reçois en avant-première les dernières tendances de l’intelligence artificielle🔥

Figure ha mostrato un nuovo video che ha per protagonista un "attore" nascosto, Helix 02, il sistema di controllo basato su una policy Vision-Language-Action che guida i suoi robot umanoidi. Nel filmato due robot F.03 sistemano una camera da letto in meno di due minuti. La scena che ha attirato più attenzione è quella del letto: i due umanoidi si mettono ai lati opposti, sollevano e stendono il piumone, correggono le pieghe e lavorano sullo stesso oggetto deformabile. Secondo Figure, i due robot usano una sola rete appresa, capace di tradurre immagini e istruzioni in movimenti. Non ci sarebbe un pianificatore condiviso, né scambio di messaggi o regia centrale. Ogni umanoide osserva la stanza dalle proprie telecamere e deduce le intenzioni dell'altro dai movimenti. Vai all'approfondimento La sequenza include anche altre azioni. Attraverso gli arti dei due robot, Helix 02 apre porte, appende un indumento, mette via cuffie su un supporto, chiude un libro, getta un rifiuto usando il pedale del cestino e spinge una sedia sotto la scrivania. Servono camminata, equilibrio, mani e lettura continua dell'ambiente. La collaborazione intorno al letto resta il passaggio più interessante. Un piumone non ha una posa fissa, si piega, scivola e cambia forma dopo ogni tiro. Quando un robot modifica la tensione del tessuto, anche il compito dell'altro cambia nello stesso istante. Come sempre in questi casi, la demo però mostra una camera specifica e non chiarisce quante prove siano state necessarie, o quanto cambi l’esito con oggetti disposti diversamente o con arredi non presenti nei dati di addestramento. Cioè, in sostanza, se i robot F.03 con Helix 02 riuscirebbero a destreggiarsi in qualsiasi altro ambiente. © riproduzione riservata Copyright © 2026 DDay.it - Scripta Manent servizi editoriali srl - Tutti i diritti sono riservati - P.IVA 11967100154

Figure AI pubblica un aggiornamento cruciale sulla rampa di produzione del suo robot umanoide di terza generazione, il F.03, nello stabilimento BotQ. Il video, caricato il 29 aprile 2026, dura circa 2 minuti e 53 secondi e mostra le impressionanti conquiste industriali raggiunte dall’azienda. In soli 120 giorni, Figure ha scalato la produzione di 24 volte, passando da 1 robot al giorno a 1 robot all’ora. Questa crescita esponenziale dimostra la maturità del processo manifatturiero, passando da prototipi artigianali a una linea di produzione automatizzata e scalabile. Figure AI, fondata nel 2022 da Brett Adcock e con sede a Sunnyvale in California, si posiziona come leader nella robotica AI per creare umanoidi autonomi commercialmente viabili. BotQ rappresenta l’impianto manifatturiero dedicato di Figure, progettato per alte volumi senza dipendere da fornitori esterni, garantendo controllo su qualità e iterazioni rapide. Questa settimana, l’azienda produrrà 55 robot F.03, confermando l’obiettivo di 12.000 unità annue sulla prima linea, con un target di 100.000 robot in quattro anni. La transizione a processi come fusione sotto pressione, stampaggio a iniezione e timbratura ha ridotto drasticamente i costi unitari, rendendo il F.03 pronto per la scala globale. Il F.03 è stato ridisegnato da zero per Helix, il modello vision-language-action di Figure, integrando percezione, ragionamento e controllo in un unico cervello generalista. Con un design più morbido e testurizzato, mani dexterous a 20 gradi di libertà e batteria integrata nel torso con densità energetica aumentata del 94% rispetto alle generazioni precedenti, il robot è ottimizzato per compiti domestici come lavare piatti o pulire. Figure mira a portare questi umanoidi in ogni casa, espandendo le capacità umane attraverso AI avanzata e produzione di massa. Sfruttando offload dati ad alta velocità, flotte di F.03 possono caricare terabyte per apprendimento continuo, rivoluzionando fabbriche e residenze. Il video non è solo un annuncio, ma una vetrina sul progresso di Figure verso la produzione seriale, evidenziando l’impegno per supply chain interne e innovazione hardware. Rilasciato in un momento di accelerazione manifatturiera, rafforza la posizione di Figure nel panorama competitivo della robotica, con focus su sicurezza, costo-efficacia e autonomia reale. Questa milestone a BotQ segnala che i robot umanoidi non sono più prototipi da laboratorio, ma prodotti pronti per il deployment massivo. Figure AI sta trasformando l’industria robotica, puntando a umanoidi versatili per case e fabbriche entro i prossimi anni. Il tuo indirizzo email non sarà pubblicato. I campi obbligatori sono contrassegnati * Commento * Nome * Email * Sito web Salva il mio nome, email e sito web in questo browser per la prossima volta che commento. Il video è la registrazione del keynote tenuto da Jensen Huang, fondatore e CEO di NVIDIA, all’NVIDIA GTC Taipei 2026, svoltosi in coincidenza con il COMPUTEX di Taipei il 31 maggio 2026 presso il Taipei Music Center. Si tratta di uno degli eventi tecnologici più attesi dell’anno, nel corso del quale Huang ha annunciato alcune […] Il video pubblicato dall’archivio della BBC è un affascinante documento storico tratto dalla trasmissione Micro Live, originariamente andata in onda su BBC Two il 12 dicembre 1986. In poco più di sei minuti, la giornalista Lesley Judd compie un viaggio dall’Inghilterra all’aeroporto di Schiphol, in Olanda, per dimostrare al pubblico televisivo dell’epoca come fosse possibile […] Il video pubblicato da OpenAI il 22 maggio 2026 presenta il lancio in preview di una nuova esperienza di finanza personale integrata in ChatGPT, disponibile per gli utenti Pro negli Stati Uniti. Connessione sicura dei conti finanziari Gli utenti possono ora collegare in modo sicuro i propri conti bancari e finanziari a ChatGPT attraverso Plaid, […] Google I/O 2026 viene presentato come il momento in cui l’azienda fa un salto netto verso una “era agentica”, in cui l’intelligenza artificiale non si limita a rispondere ma pianifica, decide e agisce nel mondo digitale degli utenti.Il keynote mette al centro Gemini come piattaforma unificata che alimenta prodotti, servizi e dispositivi, con l’obiettivo dichiarato […] Cos’è Googlebook Googlebook è il nuovo nome di una categoria di laptop che Google ha presentato recentemente nell’ambito dell’Android Show 2026, posizionandolo come il primo portatile progettato da cima a fondo per Gemini Intelligence. Non è un semplice Chromebook con un’assistente AI in più, ma un sistema operativo‑hardware pensato per integrare l’intelligenza artificiale direttamente nel […] L’Opening Keynote di Code with Claude 2026 rappresenta il momento inaugurale della conferenza organizzata da Anthropic per mostrare l’evoluzione di Claude nel lavoro degli sviluppatori, dei team tecnici e delle aziende che stanno integrando agenti AI nei propri processi. La sessione ufficiale si è tenuta a San Francisco il 6 maggio 2026, dalle 09:00 alle […] Il video presenta un’eccezionale trasformazione di una iconica Ford Mustang del 1966 in un veicolo completamente elettrico, un progetto ambizioso portato a termine da Calimotive Auto Recycling a Sacramento, California. Questa build, durata due anni, fonde il fascino vintage della muscle car americana con le tecnologie all’avanguardia di Tesla, creando un’auto unica che mantiene l’estetica […] Ubuntu 26.04 LTS, nome in codice Resolute Raccoon, è stata rilasciata il 23 aprile 2026 da Canonical come undicesima release long-term support di Ubuntu. È pensata per un uso professionale e di produzione, con un forte accento su affidabilità, sicurezza e supporto ai carichi di lavoro moderni, in particolare nell’ambito dell’intelligenza artificiale e dell’infrastruttura cloud. […] Figure AI pubblica un aggiornamento cruciale sulla rampa di produzione del suo robot umanoide di terza generazione, il F.03, nello stabilimento BotQ. Il video, caricato il 29 aprile 2026, dura circa 2 minuti e 53 secondi e mostra le impressionanti conquiste industriali raggiunte dall’azienda. In soli 120 giorni, Figure ha scalato la produzione di 24 […] Brick Machines ha realizzato una LEGO Coffee Factory capace di preparare il caffè su comando tramite smartphone, trasformando una semplice costruzione in una macchina funzionante e scenografica. Il progetto unisce creatività, automazione e spirito da laboratorio, con un risultato pensato sia per stupire sia per essere davvero utile. Come funziona la macchina La struttura utilizza […] Il video racconta Claude Design come una piattaforma in cui l’utente non si limita a descrivere un’idea, ma la trasforma in un progetto visivo vero e proprio attraverso una conversazione con Claude. Il messaggio più importante è che la progettazione viene resa più immediata, perché l’intelligenza artificiale costruisce una prima versione del lavoro e poi […] Il video “Miniature Mountain Magic: A Tilt-Shift Journey through Four Seasons in the Alps” mostra le Alpi bavaresi in una chiave visiva giocosa e spettacolare, trasformando il paesaggio in un mondo che sembra in miniatura. È un lavoro di Joerg Daiber per il progetto Little Big World, costruito con riprese aeree, time-lapse e tecnica tilt-shift. […] VOXmail è una piattaforma italiana dedicata all'invio di newsletter e all'email marketing, attiva dal 2008 e sviluppata interamente in Italia da Void Labs. Ogni giorno, migliaia di utenti si affidano a questo servizio per raggiungere i propri iscritti con comunicazioni efficaci e profession [...] FlowSpeech è una piattaforma di sintesi vocale basata sull'intelligenza artificiale che trasforma testi, documenti e immagini in audio professionale dall'intonazione naturale, con controllo avanzato delle emozioni, delle pause e dello stile narrativo. Che cos'è FlowSpeech e Come Funzion [...] PicPocket è un’app pensata per condividere e organizzare foto e video con amici, famiglia e gruppi in modo semplice, veloce e ordinato (mette a disposizione spazio per 2000 foto gratis) Sostituisce allegati email e link complicati con uno spazio condiviso che assomiglia a una chat, ma co [...] Cherri (pronunciato "cherry") è un linguaggio di programmazione open source dedicato a Siri Shortcuts, progettato per compilare direttamente uno Shortcut valido e firmato, pronto per essere eseguito su tutti i dispositivi Apple. Che Cos'è Cherri e Qual È il Suo Obiettivo Principale [...] Questo Sfondo dedicato al logo Apple per il WWDC26. Ottimo per iphone ha una risoluzione di 1206 x 2672.. E' appartenente alla categoria Tecnologia ed è reperibile all'interno dei wallpapers della pagina Apple Download Wallpaper L'immagine è ottimizzata per essere utilizzata su smartphone [...] Il video è la registrazione del keynote tenuto da Jensen Huang, fondatore e CEO di NVIDIA, all'NVIDIA GTC Taipei 2026, svoltosi in coincidenza con il COMPUTEX di Taipei il 31 maggio 2026 presso il Taipei Music Center. Si tratta di uno degli eventi tecnologici più attesi dell'anno, nel corso de [...]

Quando Figure AI ha pubblicato il video Helix-02 Bedroom Tidy, mostrando due robot umanoidi Figure 03 rifare il letto e sistemare i vestiti in una camera, l’industria della robotica ha alzato le antenne. Ma mentre tutti guardavano a ovest, dall’altra parte del mondo stava emergendo con discrezione un protagonista cinese capace di fare molto di più che girare video dimostrativi: OneRobotics. Fondata nel 2015 a Shenzhen da due laureati dell’Harbin Institute of Technology, la società è già nota a milioni di persone per la sua linea SwitchBot di dispositivi per la domotica, tra cui apricurtine intelligenti e serrature connesse. Negli ultimi anni, però, ha compiuto un salto qualitativo significativo, espandendosi nella robotica domestica basata su intelligenza artificiale con un’architettura proprietaria chiamata “One Brain, Multiple Embodiments”, il cui cuore è il modello AI OneModel, capace di condividere capacità e continuare ad evolversi tra diverse tipologie di robot. I numeri raccontano una crescita solida: il fatturato principale è passato da 275 milioni di yuan nel 2022 a 610 milioni di yuan nel 2024, con il Giappone che ha rappresentato il 68% dei ricavi nel primo semestre 2025. Non a caso, la televisione pubblica giapponese NHK ha dedicato un’intervista speciale all’azienda poco dopo la pubblicazione del video di Figure AI, concentrandosi proprio sulla dimostrazione live del robot onero H1 che, in un ambiente domestico reale, ha eseguito il flusso completo di riconoscimento degli indumenti, presa e inserimento nel cestello della lavatrice, senza scenografie costruite ad arte. Il 30 dicembre 2025, OneRobotics è approdata sul Main Board della Borsa di Hong Kong con il codice 06600.HK, diventando la prima società al mondo quotata in borsa focalizzata sui robot domestici con intelligenza artificiale embodied. L’IPO ha raccolto circa 1,64 miliardi di dollari di Hong Kong (circa 188 milioni di euro), e la capitalizzazione di mercato, già a inizio gennaio 2026, aveva superato i 23 miliardi di HKD, pari a circa 2,7 miliardi di euro. Seleziona SmartDomotica.it come fonte preferita su Google L’approccio “una sola mente, molteplici forme” si traduce concretamente in tre linee di prodotto che coprono altrettanti scenari domestici fondamentali. Il robot compagno Kata Friends si rivolge all’interazione e alla compagnia intelligente, Acemate entra nell’ambito dello sport e del benessere, mentre onero H1 affronta i servizi domestici veri e propri. Tutti condividono lo stesso cervello artificiale, con il vantaggio che ogni dato raccolto in un contesto arricchisce l’intera piattaforma. Acemate è già diventato un caso a sé: classificato come il primo robot da tennis con AI al mondo, integra visione artificiale, interazione ad alta dinamica e decision-making in tempo reale per muoversi autonomamente e rispondere ai colpi. Questa innovazione gli ha valso un posto nella celebre lista delle migliori invenzioni 2025 del Time, lo stesso riconoscimento che ha portato anche Figure 03 sotto i riflettori. Una coincidenza che dice molto sulla direzione in cui si sta muovendo l’industria. Sul fronte delle infrastrutture dati, a inizio 2026 OneRobotics ha vinto una gara pubblica a Shenzhen per la costruzione di un “Embodied Intelligence Data Full-Chain Service Center”, un contratto dal valore di 44,95 milioni di yuan (circa 5,6 milioni di euro). Il progetto prevede il dispiegamento di unità onero H1 a doppio braccio mobile, terminali di acquisizione dati UMI e sistemi indossabili per la teleoperazione, con scenari applicativi che spaziano dall’assistenza agli anziani al retail, fino alla ricerca scientifica. Si tratta di un passaggio che segna l’evoluzione dell’azienda da produttore di robot commerciali a vero e proprio fornitore di infrastrutture per la raccolta e l’addestramento dei dati, chiudendo il cerchio tra corpi fisici, scenari reali e miglioramento continuo dei modelli. Con prodotti distribuiti in oltre 90 Paesi e regioni e più di 3,6 milioni di famiglie già servite nel mondo, OneRobotics non parte da zero sul fronte della penetrazione commerciale. E proprio qui risiede forse il vantaggio competitivo più difficile da replicare: non la capacità di girare video spettacolari in ambienti controllati, ma quella di accumulare dati reali da case reali, dove le variabili sono infinite e la complessità non si può simulare in laboratorio. Seleziona SmartDomotica.it come fonte preferita su Google

Figure AI ha recentemente svelato un avanzato sistema di controllo chiamato Helix 02, progettato per gestire i suoi robot umanoidi. Questo sistema è in grado di guidare i robot senza la necessità di un pianificatore centralizzato, permettendo così una collaborazione fluida e autonoma, come dimostrato in un video che ha catturato l’attenzione del pubblico. Nel video, due robot F.03 sono stati filmati mentre sistemano una camera da letto in meno di due minuti. La scena più impressionante è quella in cui i robot collaborano per sistemare un piumone. Entrambi si posizionano ai lati opposti del letto, sollevano e stendono il piumone, correggendo le pieghe e adattandosi ai movimenti reciproci. Questa azione evidenzia non solo la loro capacità di lavorare insieme, ma anche l’efficacia del sistema Helix 02 nel tradurre istruzioni visive in movimenti coordinati. Una delle caratteristiche distintive di Helix 02 è l’assenza di una regia centrale o di uno scambio di messaggi tra i robot. Ognuno di essi utilizza le proprie telecamere per osservare l’ambiente e dedurre le intenzioni dell’altro in base ai movimenti. Questo approccio decentralizzato consente ai robot di adattarsi rapidamente ai cambiamenti nell’ambiente, rendendoli più versatili e reattivi. Il sistema Helix 02 integra una “memoria muscolare”, che consente ai robot di apprendere e migliorare le proprie abilità nel tempo. Nel video, i robot non solo sistemano il piumone, ma eseguono anche una serie di altre azioni, come aprire porte, appendere indumenti e gettare rifiuti. Tuttavia, la demo solleva interrogativi sulla versatilità di questi robot. Non è chiaro quante prove siano state necessarie per raggiungere tali risultati e come si comporterebbero in ambienti con arredi o oggetti non presenti durante l’addestramento. La tecnologia sviluppata da Figure AI rappresenta un passo significativo verso il futuro della robotica. La capacità di lavorare in modo autonomo e collaborativo potrebbe avere applicazioni in vari settori, dall’assistenza domestica alla logistica. Tuttavia, è fondamentale continuare a esplorare le possibilità e i limiti di questi sistemi per comprendere appieno il loro potenziale e le sfide che si presenteranno. © 2003 - 2025 Mr. Webmaster ® è un marchio registrato.E' vietata ogni forma di riproduzione. Un progetto a cura di IKIweb Internet Media S.r.l. - P.IVA: 02848390122 - NREA: VA-294824 - Cap. soc. 10.000 Eu i.v. - Sede legale: Via Varzi 6, Busto A. (VA) - Sede operativa: Vicolo dell'Assunta 5, Busto A. (VA) Gestisci le preferenze pubblicitarie

Décidément, l’heure est à la phase de la production de masse pour les robots humanoïdes. La firme américaine Figure AI franchit une étape industrielle majeure en annonçant avoir multiplié par 24 son rythme de production en moins de quatre mois, passant d’un robot par jour à un robot par heure pour son modèle Figure 03 ! Cette montée en cadence s’appuie sur BotQ, l’usine conçue par Figure AI pour fabriquer ses humanoïdes à plus grande échelle. L’entreprise indique avoir déjà produit plus de 350 robots Figure 03 et démontré le cycle nécessaire à ses objectifs de production : « Nous avons réussi à démontrer le temps de cycle d’un robot par heure nécessaire à nos objectifs de production du Figure 03. »  Pour y parvenir, Figure AI a mis en place des lignes dédiées aux modules critiques du robot, pilotées par un logiciel interne de gestion de production. Plus de 150 postes de travail connectés seraient désormais utilisés dans cette structure industrielle. Outre cette production de masse, Figure AI dispose aussi d’une unité dédiée à l’entrainement des robots située dans son quartier général : dans ce lieu unique, on peut voir des robots déambuler dans les couloirs ou répéter des gestes précis, le tout donnant l’impression d’être littéralement projeté dans une version réelle de la série Westworld (voir vidéo ci-dessous). La production de centaines de robots permet ainsi Figure AI d’accumuler davantage de données terrain (et plus rapidement) pour améliorer Helix, son système d’intelligence artificielle destiné à piloter les tâches physiques du quotidien. La société affirme aussi avoir dépassé les 9 000 actionneurs produits, avec plus de dix variantes de composants. Figure AI espère ainsi réduire les coûts, fiabiliser les machines et préparer leur déploiement dans les entreprises… avant d’envisager la vente aux particuliers. Signaler une erreur dans le texte Merci de nous avoir signalé l'erreur, nous allons corriger cela rapidement. Δ Nous nous réservons le droit de supprimer les commentaires qui ne respectent pas ces règles Meta a annoncé aujourd’hui de nouveaux garde-fous sur Instagram, Facebook et Messenger pour limiter l’exposition... SFR annonce la disponibilité de « SFR Navigation Protégée », une nouvelle protection de... Anthropic a déposé son projet d’introduction en Bourse, ouvrant une nouvelle phase pour l’un des groupes les plus suivis de... Le premier indice de qualité de MoffettNathanson bouscule la hiérarchie habituelle du streaming en plaçant Apple TV devant Netflix.... Amazon annonce les dates pour le Prime Day 2026, son événement shopping avec plein de réductions : ce sera du 23 juin à... Météo Météo Musique Musique Jeux Musique Musique Divertissement Utilitaires Jeux Thriller Thriller Horreur Thriller Thriller Drame Action et aventure Comédie 2 Jun. 2026 • 16:00 2 Jun. 2026 • 15:11 2 Jun. 2026 • 13:33 2 Jun. 2026 • 11:26 Actualité High-Tech, Culture Geek et comparateur de prix Recherchez le meilleur prix des produits Hi-tech Recherchez des articles sur le site

Figure AI ha messo un robot umanoide contro un lavoratore reale in una sfida di 10 ore. Il risultato finale mostra quanto la distanza si sia ridotta. Pubblicato: 18 Maggio 2026 Biagio Petronaci Tech Editor Scrive di tecnologia e innovazione digitale analizzando trend e impatti socio-culturali. Per anni, il racconto sull’automazione industriale è stato accompagnato dalla stessa idea: i robot avrebbero superato rapidamente gli esseri umani nei lavori ripetitivi. La prova organizzata da Figure AI restituisce invece uno scenario più complesso. Dopo dieci ore consecutive di lavoro in magazzino, il robot umanoide F.03 non è riuscito a battere un dipendente umano in una gara di smistamento pacchi. Il distacco finale è stato minimo, ma sufficiente a lasciare un dato concreto: almeno oggi, nelle attività fisiche reali, l’uomo conserva ancora un vantaggio. La sfida è stata trasmessa in diretta streaming e ha attirato molta attenzione online, anche perché rappresenta uno dei test pubblici più chiari sul livello raggiunto dalla robotica umanoide applicata alla logistica. Il test prevedeva un compito semplice da descrivere, ma difficile da sostenere per ore senza rallentamenti. Il robot e il concorrente umano dovevano individuare il pacco, riconoscere il codice a barre, afferrarlo e posizionarlo correttamente su un nastro trasportatore con il barcode rivolto verso il basso. Da una parte c’era il robot F.03 di Figure AI, dall’altra Aime, intern dell’azienda. Entrambi hanno lavorato per dieci ore consecutive mantenendo la stessa routine operativa. Alla fine della prova, il lavoratore umano ha elaborato 12.924 pacchi contro i 12.732 completati dal robot. Anche la differenza media per singola operazione è stata minima: 2,79 secondi per pacco contro 2,83. Il dato più interessante è proprio questo. Non si tratta più di una macchina lontana dalle prestazioni umane, ma di un sistema ormai abbastanza vicino da rendere il confronto credibile anche in un contesto operativo reale. Durante la competizione, Aime ha effettuato la pausa pranzo e i momenti di riposo previsti dalle normative sul lavoro. Il robot, invece, ha continuato a operare senza interruzioni. Secondo quanto dichiarato dal CEO Brett Adcock, il dipendente avrebbe concluso la giornata con un forte affaticamento fisico e problemi all’avambraccio dopo ore di movimenti ripetuti. È un elemento che cambia il significato del risultato finale. In una singola giornata, il vantaggio umano è rimasto intatto, ma la resistenza sul lungo periodo continua a essere uno dei principali punti di forza delle macchine. Figure AI sostiene infatti che i propri sistemi siano in grado di lavorare per turni molto più lunghi senza cali fisici evidenti. Nei test pubblicati dalla società, i robot avrebbero continuato a smistare pacchi anche oltre la durata della sfida pubblica. La logistica resta uno dei principali obiettivi della robotica umanoide proprio per la natura ripetitiva delle attività di magazzino. Figure AI punta a dimostrare che i suoi robot possano lavorare accanto agli esseri umani, ma secondo alcuni esperti la tecnologia non sarebbe ancora pronta per un’adozione industriale davvero ampia e stabile. La sfida organizzata da Figure AI mette in evidenza un paradosso sempre più evidente. Mentre molte aziende sostengono che l’intelligenza artificiale automatizzerà rapidamente i lavori d’ufficio, i robot umanoidi non riescono ancora a superare gli esseri umani nelle attività fisiche reali. Nonostante i progressi della robotica, il vantaggio umano nei contesti operativi concreti resta tangibile. Allo stesso tempo, però, il risultato ottenuto da F.03 rimane significativo proprio per la distanza ormai ridotta rispetto alle prestazioni di un lavoratore umano. FAQ Chi ha vinto la sfida tra F.03 e l'umano? L'intern Aime ha smistato 12.924 pacchi contro 12.732 del robot F.03, risultando vincitore nella giornata. Qual era il compito della prova? Individuare il pacco, leggere il codice a barre, afferrarlo e posizionarlo con il barcode rivolto verso il basso. Quanto era la differenza media per pacco? La differenza media per singola operazione è stata minima: 2,79 s per l'umano contro 2,83 s per il robot. Il robot ha fatto pause durante la prova? No, il robot ha operato senza interruzioni mentre l'umano ha osservato le pause previste dalle normative. Perché i magazzini sono adatti ai robot umanoidi? La logistica implica attività ripetitive e turni lunghi, dove la resistenza e la continuità delle macchine sono un vantaggio. L'intern Aime ha smistato 12.924 pacchi contro 12.732 del robot F.03, risultando vincitore nella giornata. Individuare il pacco, leggere il codice a barre, afferrarlo e posizionarlo con il barcode rivolto verso il basso. La differenza media per singola operazione è stata minima: 2,79 s per l'umano contro 2,83 s per il robot. No, il robot ha operato senza interruzioni mentre l'umano ha osservato le pause previste dalle normative. La logistica implica attività ripetitive e turni lunghi, dove la resistenza e la continuità delle macchine sono un vantaggio. Come funzionerà il Custom Feed di YouTube? Samsung, la soundbar top oggi costa pochissimo: va comprata subito Smart TV Panasonic, prezzo mai visto prima: solo oggi costa la metà Galaxy A56, con lo sconto di oggi il prezzo crolla al minimo storico Come potrebbe essere fatta una forma di vita aliena? Ecco qualche ipotesi Amazon anticipa il Prime Day 2026: offerte esclusive e iniziative dedicate al Sud Italia Pop-up "polyfill.io" blocca le smart TV Samsung: cos'è e come risolvere subito Nvidia vuole reinventare il PC: il nuovo chip RTX Spark che sfida Intel e Apple Investimenti, incentivi e strategie per il 2026 © Italiaonline S.p.A. 2026Direzione e coordinamento di Libero Acquisition S.á r.l.P. IVA 03970540963

Teknoloji ve bilim dünyasını seven ve takip etmekten büyük zevk alan Metin, öğrendiklerini ise DonanımHaber okuyucuları ile paylaşır. Tam Boyutta Gör ABD merkezli robotik şirketi Figure AI, BotQ üretim tesisinde insansı robot modeli Figure 03 için üretim hızını günde bir adet seviyesinden saatte bir adet seviyesine çıkararak yalnızca dört ay içinde 24 katlık artış elde etti. Bu gelişme, şirketin prototip aşamasından seri üretime geçtiğini net biçimde ortaya koyuyor. Üretim artışı, şirketin geliştirdiği özel yazılım altyapısı ve 150’den fazla birbirine bağlı iş istasyonundan oluşan üretim hattı sayesinde mümkün oldu. Figure, bu ölçekleme süreciyle birlikte bugüne kadar 350’den fazla robot teslimatı gerçekleştirdiğini açıkladı. Şirketin hedefi ise yıllık 12.000 robot üretim kapasitesine ulaşmak ve bu kapasiteyi daha da yukarı taşımak. Şirket, sadece bu hafta 55 adet insansı robot üreteceğini de açıkladı. Tam Boyutta Gör Figure, üretim verimliliğini artırmak için tedarik zincirinde kalite standartlarını sıkılaştırdı ve üretim sürecine 50’den fazla ara kontrol noktası ekledi. Bu yaklaşımın sonucunda nihai üretim hattında ilk denemede başarı oranı yüzde 80’in üzerine çıktı. Özellikle kritik bileşenlerde dikkat çekici sonuçlar elde edildi. Batarya üretiminde verimlilik yüzde 99,3 seviyesine ulaşırken toplamda 9.000’den fazla aktüatör üretildi. Şirket ayrıca 500’ün üzerinde robotun sevkiyatını tamamladı. Bununla birlikte üretilen her bir robot, üretim sonrası kapsamlı test süreçlerinden geçiriliyor. 80’den fazla fonksiyonel test, erken arızaları önlemek amacıyla uygulanıyor. Bu testler arasında çömelme ve koşu gibi fiziksel dayanıklılık senaryoları da yer alıyor. Şirketin robot filosu büyüdükçe robotlarda kullanılan yapay zeka modeli Helix için de daha fazla ölçekli veri toplanmış oluyor. Bu da robotların otonom yeteneklerinin geliştirilmesine doğrudan katkı sağlıyor. Robotlarla veri merkezi inşa edecek şirket geliyor 1 ay önce eklendi Ayrıca Figure, robot filosunu yönetmek için uzaktan güncelleme (OTA), servis ve filo yönetim sistemleri geliştirdi. Bu sistemler sayesinde sahadaki robotlar sürekli izlenebiliyor, güncellenebiliyor ve elde edilen geri bildirimler doğrudan geliştirme süreçlerine aktarılabiliyor. Şirket, üretim tarafındaki ilerlemelerin yanı sıra yapay zeka modelinde de önemli bir güncelleme duyurdu. Helix System 0 (S0) adlı modelin yeni versiyonu, robotlara çevreyi algılayarak hareket etme yeteneği kazandırıyor. Önceki sürüm yalnızca robotun kendi eklem hareketlerini ve pozisyonunu algılayan proprioseptif verilere dayanıyordu. Bu durum, merdiven veya engebeli zemin gibi karmaşık ortamlarda hareket kabiliyetini sınırlıyordu. Yeni güncellemeyle birlikte robotlar artık stereo kameralar aracılığıyla elde edilen görsel verileri kullanarak çevrenin üç boyutlu haritasını oluşturabiliyor. Bu sayede robot, bulunduğu ortamı hem “hissedebiliyor” hem de “görebiliyor”. Bu sistemin simülasyon ortamında farklı ve rastgele zemin koşullarında pekiştirmeli öğrenme yöntemiyle uçtan uca eğitildiğinin altı çiziliyor. En dikkat çekici noktalardan biri ise bu öğrenilmiş davranışların, ek bir kalibrasyon gerektirmeden doğrudan gerçek dünyaya aktarılabilmesi oldu. Sonuç olarak Figure 03 robotları artık merdivenleri çıkabilen, farklı zeminlerde dengeli hareket edebilen ve değişken ışık koşullarında stabil performans gösterebilen bir yapıya kavuştu. İnsansı robot Atlas bu kez buzdolabı taşıdı {{Description}} https://www.amazon.com.tr/dp/B0BM62VSHP https://www.amazon.com.tr/dp/B0DT1KLV4H https://www.amazon.com.tr/dp/B0CSNGM6D6 https://www.amazon.com.tr/dp/B09XBTQZF2 https://app.hb.biz/dB52DXc6OZqj https://store.steampowered.com/app/858710/Gravity_Circuit

Published on 01.Jun.2026 in Vol 28 (2026) Authors of this article: 1Department of Educational Sciences, University of Catania, Catania, Sicily, Italy 2Kent and Medway Medical School, University of Kent and Canterbury Christ Church University, Canterbury, England, United Kingdom 3Department of Sport Science, School of Science and Technology, Nottingham Trent University, Nottingham, England, United Kingdom 4Department of Computing, Sheffield Hallam University, Sheffield, England, United Kingdom 5School of Health Sciences, University of Southampton, Building 67, Highfield Campus, University Road, Southampton, England, United Kingdom Daniele Magistro, PhD Background: Maintaining cognitive efficiency and independence is a central goal of healthy aging. Socially assistive robots (SARs) are increasingly proposed as scalable digital health solutions to support daily activities in older adults and to facilitate aging-in-place. However, concerns remain regarding whether robot-mediated assistance reduces or inadvertently increases cognitive load, potentially undermining usability, user acceptance, and long-term real-world adoption, particularly in aging populations. Objective: This study aimed to examine how robot-assisted (human-robot interaction [HRI]) and human-assisted (human-human interaction [HHI]) support influences cognitive load during task performance in younger and older adults. A multimodal assessment framework integrating behavioral, subjective, and physiological measures was used to identify age-related differences in cognitive effort and stress associated with different forms of assistance. Methods: A total of 60 healthy adults (30 younger adults: mean age 34.8, SD 10.1 years; and 30 older adults: mean age 72.3, SD 5.5 years) completed a modified Trail Making Test under 7 within-subject conditions: independent performance (baseline), 3 robot-assisted conditions, and 3 human-assisted conditions, each corresponding to low, medium, and high cognitive load levels. Performance accuracy and completion time were recorded as behavioral indicators. Perceived cognitive load was assessed using the National Aeronautics and Space Administration Task Load Index, and physiological stress was evaluated via pre- and postcondition salivary cortisol concentrations. Linear mixed-effects models were applied to examine main effects and interactions of age group, assistance type, cognitive load level, and time. Results: Significant interactions between age group and assistance type were observed for accuracy (F1, 404.53=6.50; P=.01) and perceived cognitive load (F1, 403.45=4.58; P=.03). Older adults demonstrated lower accuracy and higher perceived cognitive load during robot-assisted conditions compared with human-assisted conditions, whereas no such differences were observed in younger adults. Across age groups, human assistance improved performance at low and medium cognitive load levels. Physiological analysis revealed a significant age×assistance× time interaction (F1, 156=5.16; P=.02), with older adults showing increased posttask cortisol concentrations during robot-assisted interaction, indicating higher physiological stress. Conclusions: While both human and robotic assistance enhanced task performance relative to independent completion, the type of support critically shaped cognitive load responses in older adults. Robot-assisted interaction was associated with increased behavioral errors, higher perceived workload, and elevated physiological stress, suggesting that current SAR implementations may impose additional extraneous cognitive load in older users. These findings highlight the importance of designing adaptive, age-sensitive digital assistive systems that minimize cognitive burden through simplified interaction, responsive pacing, and multimodal support. Multimodal cognitive load assessment provides a valuable framework for optimizing the usability and effectiveness of assistive digital health technologies for aging populations. Maintaining independence and cognitive efficiency in daily life constitutes a central objective in the promotion of active and healthy aging [1]. The ability to manage daily activities with adequate support has become a key focus across multiple disciplines, particularly in light of the projected rise in the global mean age over the coming decades [2]. In this context, health and social care systems are increasingly exploring digital health technologies to support autonomy, reduce care burden, and enable aging in place, underscoring the need for scalable and evidence-based interventions aimed at enhancing quality of life among older adults. Technological progress has played a pivotal role in pursuing this goal. Intelligent systems can now be embedded within domestic environments, and embodied agents may be programmed to interact with users to accomplish shared objectives [3]. Within this framework, socially assistive robotics (SAR) leverages human-robot interaction (HRI) to facilitate such activities and to foster autonomy in everyday life [4]. SAR systems are increasingly positioned as interactive digital health interventions, integrating social interaction, guidance, and task support within real-world settings. However, recent evidence indicates that enabling communication with an artificial agent alone is insufficient to ensure effective interaction, particularly in vulnerable groups such as older adults [5]. Beyond demonstrating performance benefits, digital assistive technologies must also be evaluated in terms of usability, acceptability, and sustainability of long-term adoption [6,7]. Interventions that improve task execution but simultaneously increase cognitive or physiological strain may face significant barriers to real-world implementation. From a cognitive ergonomics perspective, cognitive load (CL) represents a critical determinant of HRI effectiveness and of the usability and sustainability of digital health technologies [8]. Conceptually, CL is defined as the amount of mental resources required to perform a given task [9]. Because cognitive resources are inherently limited, individuals must distribute them efficiently across competing activities [10]. Importantly, CL is not a unitary construct but rather comprises three interrelated components: intrinsic, extraneous, and germane load. Intrinsic load stems from the inherent complexity of the task and the volume of information to be processed. Extraneous load arises from the presentation format and the presence of external distractions, both of which can be mitigated through effective task design. Germane load, instead, pertains to the cognitive processes that foster learning and schema construction. Recent theoretical models have proposed integrating intrinsic and germane load, conceptualizing intrinsic and extraneous load as the two principal dimensions of CL theory [11-13]. In applied digital health contexts, extraneous CL introduced by interface design is particularly problematic, as it may negate the intended supportive function of assistive technologies. A growing body of research indicates that older adults face greater challenges in managing CL, thereby affecting their functional autonomy [14]. Such difficulties are often attributed to age-related cognitive decline, which manifests as increased fatigue and reduced efficiency in completing daily tasks [15]. These challenges extend to the use of technology, where the additional cognitive demands associated with complex interfaces further amplify CL [16,17]. Moreover, older adults’ attitudes toward advanced technologies are shaped not only by sociocultural factors but also by individual cognitive needs and the degree of system personalization [18]. Neglecting these factors’ risks imposing additional cognitive burdens and may contribute to digital exclusion among those most in need of support. Despite the growing interest in SAR, few studies have directly compared robot-mediated assistance with human assistance under controlled CL conditions. Such comparisons are essential, as assistive technologies are often implicitly framed as substitutes for human support. Furthermore, assessing technology-assisted performance without an independent baseline condition limits the ability to determine whether observed effects reflect genuine support or task facilitation, as highlighted in previous work [19]. Accordingly, it becomes essential to determine whether technology use increases users’ CL, to identify the features responsible for such increases, and to examine how these effects differ across age groups. Such insights are crucial for the design of adaptive interfaces capable of dynamically accommodating users’ cognitive capacities. Nevertheless, the existing literature presents several methodological limitations in addressing these issues. First, assessing CL during technology use without a baseline condition prevents reliable interpretation. Experimental designs should therefore include a comparison between technology-assisted and independent task performance to delineate the specific benefits and drawbacks of technological aids, a methodological framework exemplified by Varrasi et al [19]. Furthermore, studies asserting the positive impact of technology without incorporating a condition involving human support lack robustness; indeed, multiple aid conditions are required for a comprehensive evaluation. Likewise, hypothesizing user needs without conducting age comparisons constrains the development of genuine person-centered support systems. Specific sampling is thus necessary to enable between-group analyses and to characterize distinct population profiles. Finally, the reliance on a single CL measure limits construct validity, underscoring the need to use multimodal assessment approaches. CL can be quantified through a range of behavioral, subjective, and physiological indicators, each providing complementary information on the cognitive demands placed on the individual. Behavioral indices include performance-based metrics, such as accuracy, error rates, and completion time, which serve as quantifiable markers of cognitive effort [20]. Subjective indices capture self-reported perceptions of mental workload, encompassing aspects such as perceived effort, frustration, and mental demand [21]. Integrating these approaches enables a comprehensive evaluation of cognitive demands and supports the robust assessment of assistive technologies in digital health contexts. Physiological indices, including heart rate variability, pupil dilation, blood and salivary cortisol concentrations, and electroencephalographic (EEG) activity, reflect the body’s physiological responses to cognitive strain and stress [22]. Integrating these methods offers a comprehensive assessment of CL and facilitates a more accurate evaluation of task design and learning efficiency. Given these considerations, this study aimed to investigate whether, and in what ways, HRI and human-human interaction (HHI) influence CL management during task performance in younger and older adults. By incorporating an independent baseline condition and a multimodal psychometric framework, this study seeks to inform the design, evaluation, and deployment of age-sensitive digital assistive technologies. This study used a within-subject, experimental comparative design to examine differences in CL during independent, robot-assisted, and human-assisted task performance in younger and older adults. The study included 60 adult volunteers, comprising 30 younger adults (13 males, 17 females; mean age 34.8, SD 10.1 years) and 30 older adults (12 males, 18 females: mean age 72.3, SD 5.5 years), with sex distribution confirmed as balanced (χ²1=0.07; P=.79). Younger adults were aged 18‐45 years, and older adults were more than 65 years of age. In line with the study aims, age differed significantly between groups (t58=17.86; P<.001). Education levels were broadly comparable across groups. Among younger adults, 16.7% (5/30) reported high school education, 23.3% (7/30) bachelor’s, 30% (9/30) master’s, and 30% (9/30) PhD-level education. Among older adults, 33.3% (10/30) reported high school education, 30% (9/30) bachelor’s, 23.3% (7/30) master’s, and 13.3% (4/30) PhD-level education. Although younger adults showed a tendency toward higher educational attainment, this difference was not statistically significant (χ²3=4.09; P=.25). An a priori power analysis (2-tailed α=.05, 80% power) indicated that the sample size of 30 adults per group provides sufficient power to detect Cohen d∼0.74 for between-group comparisons and interaction effects (approximated as between-group differences in within-subject contrasts). Within-group paired contrasts have 80% power to detect dz∼0.53 (n=30) and dz∼0.37 when considering the full sample (N=60). These estimates represent conservative approximations relative to the mixed-effects modeling approach used in the primary analyses. All participants underwent screening to confirm good general health, as only healthy individuals were included in the study. Specifically, participants had no prior diagnosis of neurological or psychiatric disorders and reported no previous direct experience interacting with socially assistive robots. This sampling strategy was adopted to isolate age-related differences in CL responses under controlled experimental conditions while minimizing additional variability associated with clinical impairment. Accordingly, the study was conceived as an initial step toward understanding how assistance modality influences CL, with the intention of extending this line of research to more vulnerable populations in future work. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was obtained from the Nottingham Trent University Ethics Board (reference number 729) in March 2024. All participants provided written informed consent prior to participation and were informed of their right to withdraw at any time without consequence. The cognitive task used in this study was adapted from the B form of the Trail Making Test (TMT) [23]. The TMT was selected over alternative cognitive measures due to its engagement of multiple cognitive domains, including attention, visual search, executive functioning, and working memory. Consequently, the task required participants to simultaneously deploy multiple cognitive skills, closely reflecting the complexity of everyday activities. Specifically, the task used a 29.7×42 cm sheet of paper containing 52 randomly arranged circles (Figure 1). Half of the circles included English letters (A-Z), while the remaining circles contained numbers (1-26). A total of 7 task versions were created to minimize learning effects across repeated conditions. These versions were not intended to represent validated parallel forms of the TMT in a clinical sense; rather, they were task variants inspired by the TMT and designed to evoke comparable executive, attention, and visual search demands. For each version, the 52 circles were randomly distributed across the sheet to create distinct spatial arrangements while preserving the same basic task structure. Task versions were assigned randomly to participants across conditions, with each participant completing a different version in each condition. Objective performance was behaviorally evaluated based on the number of correctly connected pairs and the time required to complete the task (seconds). The supportive agent used in this study was the NAO 6 humanoid robot (United Robotics Group). This platform was selected for its commercial availability and relevance to real-world assistive deployments aligned with the study’s requirements, including speech production, speech recognition, and social interaction. Previous studies have shown that NAO (Aldebaran, Paris, France) functions as an SAR that is generally well-received by older and vulnerable populations [16]. The robot is 58 cm tall and can interact through speech, gestures, and facial tracking. Programming was implemented using Choregraphe software (version 2.8.7; Aldebaran) together with Python (Python Software Foundation) to enable interactive collaboration with participants in 3 of the 7 experimental conditions. The robot’s behavior was structured using block-based visual programming within a node flow interface. Interactions began with NAO tracking the participant’s face to simulate eye contact, followed by a self-introduction and delivery of initial task instructions. Speech recognition allowed the robot to repeat instructions upon participant request during an initial familiarization phase, which served as a vocal adaptation period responsive to participant feedback. After this phase, NAO did not repeat instructions to avoid disrupting task performance, although speech recognition remained active to manage pacing during the interaction. Indeed, the robot-participant interaction followed a structured and standardized dialogue logic designed to ensure consistency across participants while maintaining naturalistic turn-taking dynamics. Specifically, interaction sequences were organized into alternating robot-led instruction phases and participant execution phases. The robot presented task instructions verbally according to predefined scripts and advanced through task steps only after receiving explicit verbal confirmation from the participant (eg, “Okay” or “Done”), thereby regulating the interaction rhythm. The presentation of task instructions varied systematically according to CL conditions. In the low-load condition, the robot delivered one pair of items at a time, whereas in the medium- and high-load conditions, multiple pairs were provided sequentially within a single instruction block. The interaction pacing was therefore partially scripted, with fixed instruction sequences, but temporally modulated by participant responses, allowing limited responsiveness to individual execution speed. To ensure procedural reproducibility, the interaction logic combined fixed scripted content (instruction wording, sequence structure, and load manipulation) with controlled interaction triggers based on participant verbal feedback. A schematic representation of the interaction workflow is provided in Figure 2. The flowchart illustrates the decision nodes governing instruction delivery, response confirmation, and transition between interaction phases, providing a visual representation of the standardized dialogue protocol implemented across conditions. Moreover, to enhance transparency and reproducibility of the interaction protocol, a structured overview of the robot dialogue architecture, including interaction phases, scripted instruction templates, and response-triggered transitions, is provided in Table 1. Perceived cognitive load (PCL) was assessed using the National Aeronautics and Space Administration Task Load Index (NASA-TLX) [24], a brief, self-administered psychometric tool evaluating 6 dimensions: mental demand, physical demand, temporal demand, performance, effort, and frustration. Each dimension was rated on a 20-point scale, with higher scores indicating greater perceived strain. The NASA-TLX was chosen for its robust psychometric properties and demonstrated superiority over alternative subjective workload measures [25], making it well-suited for assessing PCL in this study. In line with the primary aim of capturing overall perceived workload during assisted task performance, the total NASA-TLX score was used as a global indicator of subjective CL. Although the instrument allows examination of individual workload dimensions, the aggregated score has been widely used to provide a comprehensive estimate of perceived task demand, particularly in applied human-technology interaction research [25]. To objectively assess physiological stress responses associated with the CL tasks, saliva samples were collected immediately before and after each condition (ie, robot-assisted and human-assisted). Samples were obtained using a synthetic swab (Salivette Cortisol, Sarstedt), instructing participants to slowly move the swab around the oral cavity for 2 minutes, in accordance with the manufacturer’s instructions. All samples were maintained on crushed ice throughout testing, which did not exceed 45 minutes. Subsequently, saliva samples were centrifuged at 1000×g for 2 minutes (Espresso Microcentrifuge, Thermo Scientific), and the supernatant was transferred into 1.5 mL Eppendorf tubes and stored at −80 °C until analysis. Quantitative determination of salivary cortisol concentrations was performed using commercially available enzyme-linked immunosorbent assay kits (Salimetrics), following the manufacturer’s recommended procedures. Salivary cortisol was included as a physiological indicator of stress-related arousal associated with cognitive effort. While cortisol does not represent a direct measure of CL, it provides complementary information regarding the neuroendocrine response to task-related strain, allowing integration with behavioral and subjective workload indicators. Moreover, salivary cortisol was sampled only before and after each assisted block, and not after the independent baseline condition, because the physiological outcome was intended to capture the cumulative response associated with the two assistance modalities rather than transient changes across individual CL levels or during unassisted performance. This choice was based on both theoretical and methodological considerations, including the temporal dynamics of the hypothalamic-pituitary-adrenal axis, as salivary cortisol typically peaks approximately 20‐40 minutes after stress onset. Accordingly, cortisol was treated as a block-level marker of stress-related arousal, and restricting sampling to the assisted conditions helped reduce participant burden while preserving the feasibility and ecological validity of the experimental protocol. Each participant completed the experimental protocol individually across seven conditions: one baseline condition performed independently, three robot-assisted conditions at varying CL levels (low, medium, and high), and three human-assisted conditions corresponding to the same CL levels. The baseline condition was systematically administered at the beginning of the session to provide a common, unassisted reference point for all participants. Within each assisted block, CL levels were presented in a fixed progressive sequence (low, medium, and high) in both the robot-assisted and human-assisted conditions. This choice was made to ensure a gradual and controlled increase in task demands, thereby minimizing potential confusion associated with abrupt changes in difficulty and maintaining a stable and predictable interaction structure, particularly for older adults. In contrast, the order of the assisted blocks (robot-assisted vs human-assisted) was fully counterbalanced across participants to control for order effects at the level of assistance modality. Both assisted conditions were administered within the same experimental session. Consequently, the precondition saliva sample for the second assisted block was collected immediately before its onset, following completion of the preceding condition and a brief transition phase required to reset the materials and prepare the subsequent task. While counterbalancing mitigates systematic sequence effects at the level of condition order, the absence of an extended washout interval implies that a residual carryover influence from the first assisted condition on the precondition cortisol value of the second block cannot be entirely excluded. The total duration of the experimental session was approximately 30 minutes. After providing informed consent, the participant was seated at a desk and presented with the first task sheet. In the baseline condition, the participant was instructed to connect alternating numbers and letters in reverse order (ie, starting from the highest number and the last letter) as quickly and accurately as possible using a marker. This task was designed to elicit CL by engaging multiple cognitive domains, including semantic memory, working memory, visual search, and processing speed. Specifically, the participant was required to recall the reverse sequence of alphanumeric items, maintain this sequence in working memory, and identify and connect the corresponding circles on the sheet. During the three robot-assisted conditions, the NAO humanoid robot acted as an interactive support agent (Figure 3). At the beginning of each assisted condition, the robot greeted the participant, introduced the task, and repeated the instructions upon request. Throughout task execution, the robot verbally provided the pairs of circles to be connected, thereby reducing reliance on semantic memory and allowing participants to focus on visual search and motor coordination. The robot advanced through the sequence based on participant responses, waiting for a verbal confirmation (eg, “Okay”) before announcing the next pair. To prevent learning effects, the sequence of pairs was randomized across trials. CL was systematically manipulated as follows: in the low-load condition (second and fifth condition), the robot presented one pair at a time; in the medium-load condition (third and sixth condition), two to four pairs were provided simultaneously; and in the high-load condition (fourth and seventh condition), five to seven pairs were delivered concurrently. The robot did not repeat any pair throughout the session. The human-assisted condition was administered by a trained experimenter who followed a standardized script designed to mirror the robot-assisted protocol in content, sequence, and CL manipulation. The experimenter delivered the instructions verbally using the same wording as the robot prompts, maintained the same load-specific structure, and waited for the same participant verbal confirmations before proceeding to the next step. A brief initial familiarization phase was also included to ensure procedural parity with the robot-assisted condition: during this phase, the experimenter presented the task instructions and allowed participants to request clarification, analogous to the vocal adaptation period implemented in the HRI condition. Following familiarization, the interaction proceeded according to the predefined script without additional instruction repetition. The only difference between conditions was the modality of delivery, with all other procedural elements held constant. A structured overview of the human-assisted dialogue, emphasizing the fidelity rules for the adherence to the HRI condition, is provided in Table 2. At the end of each condition, including baseline, robot-assisted, and human-assisted tasks, participants completed the NASA-TLX questionnaire to evaluate PCL. As described previously, saliva samples were collected at the beginning and the end of both the robot-assisted and human-assisted conditions to capture physiological markers of CL, yielding four samples per participant. For data analysis, two independent experimenters (SV and RV) assessed task performance across all conditions. Performance measures included accuracy (number of correctly connected pairs) and completion time (in seconds). Subjective workload was quantified using NASA-TLX scores, while physiological responses were evaluated by comparing pre- and postcondition cortisol concentrations across the robot- and human-assisted settings. Linear mixed-effects models were applied using restricted maximum likelihood estimation to examine differences across the primary outcome variables, including NASA-TLX scores, task performance indices, and cortisol responses. Participants were included as random intercepts to account for interindividual variability in baseline performance and physiological measures. A type III ANOVA with Satterthwaite approximation was used to assess the main effects and interactions of condition (robot-assisted vs human-assisted), CL level (baseline, low, medium, and high), and age group (younger vs older adults). Condition at varying load levels was modeled as a within-subject factor, while age group was treated as a between-subject factor. Post hoc comparisons were performed on estimated marginal means and mean differences (Mdif) to further explore significant main effects and interactions. Tukey-adjusted estimated marginal means comparisons were used to correct for multiple testing. In line with the study’s objectives, particular attention was devoted to the interaction between age group and experimental condition to clarify differential patterns of CL management and task performance across age cohorts. Mean (SD) values for accuracy (number of correctly matched pairs), completion time (in seconds), and PCL across the seven experimental conditions, as well as cortisol levels (nmol/L) measured before and after the robot-assisted and human-assisted conditions, are presented in Tables 3 and 4. aPCL: perceived cognitive load. bCL: cognitive load. aCortisol levels (nmol/L) were measured at the beginning (pre) and end (post) of the human-human interaction and human-robot interaction conditions. Regarding performance, analysis of accuracy revealed significant interactions between age group and condition (F1, 404.53=6.50; P=.01) as well as between age group and CL level (F3, 403.45=5.88; P<.001). Post hoc comparisons indicated that older adults exhibited lower accuracy in the HRI condition compared with younger adults (Mdif=1.97, SE=0.58; t106=3.34; P=.001; Figure 4A) and also relative to the human-assisted condition (Mdif=1.54, SE=0.43; t403=3.57; P<.001; Figure 4B). Additionally, older adults demonstrated significantly lower accuracy than younger adults at medium (Mdif=2.75, SE=0.73; t219=3.75; P<.001) and high CL levels (Mdif=1.64, SE=0.73; t219=2.23; P=.03), whereas no significant differences were observed at low load (Mdif=1.16, SE=0.73; t219=1.58; P=.12) or in the baseline condition (Mdif=−0.80, SE=0.73; t217=−1.09; P=.27; Figure 4C). A significant interaction between condition and CL level was also observed (F1, 403.45=5.83; P<.001). Across all participants, post hoc tests revealed higher accuracy in the human-assisted compared with the robot-assisted condition at low (Mdif=1.25, SE=0.61; t404=2.05; P=.04) and medium CL levels (Mdif=2.58, SE=0.61; t404=4.21; P<.001), whereas no significant differences emerged at high CL (Mdif=−0.79, SE=0.61; t404=−1.29; P=.19; Figure 4D). Analysis of task completion time revealed a significant interaction between age group and CL level (F3, 402.85=23.73; P<.001). Post hoc comparisons indicated that younger adults completed the task faster than older adults at baseline (Mdif=−123.73, SE=18.70; t114=−6.62; P<.001), as well as under low (Mdif=−115.95, SE=18.70; t115=−6.19; P<.001) and medium CL conditions (Mdif=−75.16, SE=18.70; t115=−4.01; P<.001). No significant difference was observed at high CL (Mdif=1.73, SE=18.70; t115=0.09; P=.92; Figure 5). Analysis of NASA-TLX scores revealed significant interactions between age group and condition (F1, 403.45=4.58; P=.03) as well as between age group and CL level (F3, 403.12=8.53; P<.001). Regarding the interaction between age group and condition, post hoc comparisons showed that older adults reported significantly higher scores during the robot-assisted condition (mean 57.60, SE 2.13) compared with the human-assisted condition (mean 54.3, SE 2.13), with an estimated Mdif of −3.29 (SE=1.16; t403=−2.83; P=.005). No significant difference was observed in younger adults (Mdif=0.25, SE=1.18; t404=0.21; P=.83), indicating that older adults experienced greater perceived CL when interacting with the robot compared with human support (Figure 6A). For the interaction between age group and CL level, post hoc analysis revealed that older adults reported significantly higher total scores at the low CL level (mean 43.8, SE 2.28) compared with younger adults (mean 37.2, SE 2.29; Mdif=−6.63, SE=3.23; t89.0=−2.05; P=.04), suggesting that older adults perceive more CL than younger adults under low-load conditions (Figure 6B). Analysis of the physiological measure of salivary cortisol concentration revealed a significant interaction between age group, condition, and time (pre vs post; F1, 156=5.16; P=.02). When analyzed separately by age group, older adults demonstrated a significant interaction between condition and time (F1, 156=10.32; P=.001), whereas younger adults showed no significant effect (F1, 156=0.01; P=.90). Post hoc comparisons indicated that older adults exhibited a significant increase in cortisol concentrations from baseline to postintervention during the HRI condition (Mdif=−5.65, SE=1.31; t156=−4.31; P=.001; Figure 7). This study aimed to investigate how HHI and HRI support the management of CL during task performance in younger and older adults, integrating behavioral, subjective, and physiological measures. By including a baseline condition without assistance and systematically varying CL levels, this research extends the literature on cognitive ergonomics in aging and technology-assisted contexts by providing an applied evaluation of assistive support modalities that are increasingly proposed as digital health solutions for aging populations. Importantly, these findings should also be interpreted within a translational digital health perspective. While assistive technologies are frequently evaluated in controlled experimental settings, their successful implementation depends on sustained usability, user acceptance, and ecological integration into daily routines. In this context, increases in CL or physiological load, even when accompanied by performance improvements, may represent critical barriers to long-term engagement and adherence. Consequently, evaluating assistive systems through a multidimensional framework that captures both performance outcomes and user burden is essential to inform scalable and sustainable deployment strategies. Consistent with previous evidence indicating age-related declines in processing speed and working memory capacity [26,27], older adults demonstrated lower accuracy and required more time to complete tasks as CL increased. These differences were most pronounced at medium- and high-load levels, supporting theoretical frameworks positing that increased task complexity disproportionately affects older adults due to age-related cognitive changes [28]. Notably, at low CL levels, accuracy differences between age groups were reduced, suggesting that minimal task demands may better align with preserved cognitive capacities in older adults, enabling performance comparable to that of younger adults. Importantly, these findings reinforce the need to evaluate assistive technologies for older adults not only under simplified or optimal conditions, but also across varying levels of cognitive demand that more closely reflect real-world task complexity. The interaction between type of assistance and CL revealed that HHI generally supported higher accuracy compared with HRI, particularly at low- and medium-load levels. This finding aligns with research suggesting that human support can more flexibly and adaptively accommodate the nuanced needs of older adults, thereby reducing extraneous CL [29]. The absence of significant accuracy differences between HRI and HHI at high CL may tentatively reflect a ceiling effect, whereby performance approached an upper limit, reducing sensitivity to differences between assistance modalities. However, this interpretation should be treated with caution, as performance saturation was not formally tested in this study. Subjective measures of PCL, assessed via NASA-TLX scores, indicated that older adults consistently reported higher CL during HRI compared with HHI, particularly under low-load conditions. This finding is especially relevant from a usability and adoption perspective, as increased perceived workload under otherwise manageable task demands may negatively influence trust, acceptance, and sustained engagement with assistive technologies. This result suggests that even minimal interactions with robotic agents can impose additional extraneous cognitive demands for older adults, potentially due to challenges in speech recognition, turn-taking, or the social-cognitive demands of interpreting robotic behavior [30]. It is important to acknowledge that certain technical constraints are inherent to current HRI systems. Features such as response latency, variability in speech recognition, and predefined interaction pacing represent realistic operational characteristics of SARs rather than limitations of the experimental design. These factors may require users to allocate additional attentional and executive resources to manage turn-taking and system feedback, thereby contributing to increased extraneous CL. Such demands may be particularly relevant for older adults, who may be more sensitive to temporal unpredictability and reduced interactional flexibility. Recognizing these constraints as ecological features of present technologies highlights key targets for future system refinement, including more robust speech processing and adaptive interaction pacing, which may ultimately improve usability and reduce cognitive burden. By contrast, younger adults showed no differences in PCL between assistance types, suggesting that HRI may be more readily accommodated by individuals with higher baseline cognitive flexibility and greater familiarity with technology [31]. Physiological measures of salivary cortisol concentrations corroborated these findings, revealing increased cognitive strain during HRI for older adults, with significant postintervention elevations observed in the HRI condition. This is consistent with prior literature demonstrating heightened physiological reactivity in older adults to tasks perceived as challenging or unfamiliar [32]. The convergence of behavioral, subjective, and physiological indicators strengthens the interpretation that HRI, in its current form, may elicit latent stress responses in older users that are not fully captured by performance metrics alone, underscoring the value of multimodal assessment frameworks in digital health evaluation. Indeed, the use of the total NASA-TLX score allowed the capture of participants’ global perception of workload across interaction modalities, providing an ecologically valid estimate of overall task burden. Complementarily, cortisol responses were interpreted as markers of physiological stress associated with cognitive effort rather than direct indicators of CL per se. The consistency of these measures supports a multidimensional interpretation of user strain during assistive interaction. The absence of similar effects in younger adults underscores the differential impact of HRI on cognitive stress across age groups and highlights the importance of integrating physiological, subjective, and performance-based measures to comprehensively assess CL. Together, these findings highlight the complex interplay among aging, CL management, and the type of assistance provided during task performance. Previous research on behavioral and cognitive-affective regulation has similarly emphasized the need for individualized and context-sensitive approaches in psychological assessment and intervention [33,34]. Within the context of assistive digital technologies, these results suggest that “one-size-fits-all” interaction models are unlikely to meet the needs of older adults with heterogeneous cognitive profiles. While robotic assistance offers scalable and consistent support, its current implementation may inadvertently increase extraneous CL for older adults, particularly in tasks requiring social-cognitive engagement. In contrast, human assistance appears to mitigate these challenges, supporting better performance and lower perceived strain under comparable task demands. From a digital health design and deployment perspective, these findings suggest that SAR systems should prioritize the minimization of extraneous CL through simplified interaction structures, improved speech recognition robustness, and adaptive pacing strategies that respond to individual user performance and stress indicators. Incorporating real-time or near–real-time indicators of CL could enable dynamic adjustment of assistance, reducing unnecessary cognitive strain and improving user experience. Moreover, hybrid models that combine robotic assistance with intermittent human support may represent a pragmatic transitional approach, particularly in early stages of adoption or in cognitively demanding tasks. These considerations are particularly relevant for large-scale implementation strategies, where technologies must remain cognitively sustainable across prolonged and repeated use in real-world environments. Beyond implications for system design and deployment, this study also offers methodological insights that strengthen the interpretation of technology-assisted CL effects. Importantly, the inclusion of an independent baseline condition in this study strengthens the interpretation of the observed effects by enabling direct comparison between assisted and unassisted task performance. This design feature allows differentiation between genuine support-related benefits and task facilitation effects that may arise simply from external guidance, irrespective of the modality of assistance. By contextualizing both human and robotic assistance against independent performance, the findings demonstrate that while assistance generally improves task outcomes, the cognitive and experiential costs associated with different support modalities vary substantially, particularly for older adults. An additional methodological strength is that the human-assisted condition was operationalized through a standardized script matched to the robot-assisted protocol in wording, sequencing, and load manipulation. This design choice reduced the risk that differences between conditions were attributable to variability in human delivery style rather than to the assistance modality itself. This comparative framework enhances internal validity and provides a more robust basis for evaluating assistive technologies as potential substitutes or complements to human support in real-world digital health contexts. The observed interaction patterns suggest that technological interventions designed for older adults should prioritize reducing extraneous CL by simplifying robot interfaces, improving speech recognition accuracy, and implementing adaptive pacing tailored to individual users. Indeed, although older adults view social robots as a potential tool to support their daily activities, our results further highlight the importance of designing such systems according to users’ specific needs [5]. Additionally, multimodal support strategies, including visual cues and simplified verbal prompts, may enhance the usability and effectiveness of HRI by aligning it more closely with the cognitive profiles of older adults. This study suffers from some limitations. For instance, the generalizability of these findings. Although the inclusion of healthy younger and older adults allowed us to examine age-related differences under controlled conditions, the results cannot be directly generalized to more vulnerable groups, such as individuals with mild cognitive impairment or other clinical conditions for whom SARs may be particularly relevant. Future studies should therefore replicate this design in clinical and functionally vulnerable populations to determine whether the observed patterns of behavioral, subjective, and physiological response are preserved, attenuated, or amplified in contexts of greater cognitive vulnerability. Then, the seven task versions were not formally tested for equivalence in difficulty. Accordingly, they should not be interpreted as psychometrically validated parallel forms of the TMT. Although random assignment of versions across participants and conditions was used to reduce systematic version effects and support internal validity, residual differences in difficulty between task versions cannot be entirely excluded. Future studies should include formal equivalence testing of task variants or adopt a fully validated parallel-form procedure if the task is to be used repeatedly across conditions. A further methodological consideration concerns the fixed progression of CL levels within each assisted block. Although this approach ensured a controlled and gradual increase in task demands and supported procedural consistency across participants, it does not allow complete separation of CL effects from potential sequence effects. Future studies may benefit from counterbalancing or randomizing load sequences, although such designs should carefully consider the potential impact of increased task-switching demands, particularly in older populations. An additional limitation concerns the interpretation of the nonsignificant differences in accuracy between conditions at high CL. Although a ceiling effect may provide a plausible explanation, this possibility was not directly tested through specific analyses of performance saturation. As such, this interpretation should be considered tentative, and future studies should include measures or analytical approaches specifically designed to detect ceiling effects. Moreover, the sequential structure of the assisted blocks should be considered when interpreting the cortisol findings. The order of robot-assisted and human-assisted conditions was fully counterbalanced, which reduced the likelihood that the results were driven by presentation order. However, because the precondition cortisol sample for the second block was collected immediately after completion of the first assisted condition, a residual carryover effect cannot be completely excluded. This limitation is particularly relevant for salivary cortisol, given its temporal dynamics and slower recovery profile. Future studies may benefit from longer recovery intervals or separate-session administration to further reduce potential carryover effects while preserving the advantages of counterbalanced designs. Another important methodological consideration is that salivary cortisol was assessed only in the assisted conditions and at the block level, whereas behavioral and subjective outcomes were collected across all load levels, including baseline. Consequently, the physiological findings should be interpreted as reflecting differences between robot-assisted and human-assisted interaction, rather than as load-specific effects or as changes relative to the independent baseline condition. This design choice was consistent with the intended use of cortisol as a cumulative marker of stress-related arousal; however, it also limits direct comparison between physiological and other outcome domains and should be considered when interpreting the multimodal findings. Future research should examine longitudinal exposure to robotic systems to determine whether increased familiarity reduces cognitive and physiological strain in older adults, potentially enhancing acceptance and efficacy. Such work is essential to distinguish short-term novelty or learning effects from stable interaction patterns that are likely to emerge during real-world deployment. Expanding physiological monitoring to include measures such as heart rate variability and electroencephalography could further elucidate the real-time cognitive dynamics associated with HRI and HHI across age groups. In conclusion, this study provides important evidence regarding CL management in aging populations within assisted task contexts, emphasizing the need to tailor support strategies to individual cognitive capacities. While HRI holds promise for facilitating aging-in-place initiatives, its effectiveness depends on careful interaction design and adaptive support mechanisms that ensure robotic assistance reduces, rather than amplifies, cognitive and physiological burden. Optimizing these systems is critical to supporting autonomy, usability, and quality of life in the context of active aging. Although both HHI and HRI improved performance compared with independent task completion, the effectiveness of assistance was strongly dependent on the type and complexity of support, particularly in older adults. Human assistance consistently supported higher accuracy and lower perceived workload, whereas robotic assistance, despite its promise for scalable and consistent support, was associated with increased perceived and physiological CL in older adults, especially under low and medium task demands. Overall, these findings highlight the importance of designing adaptive, age-sensitive digital assistive systems that minimize cognitive burden through simplified interaction, responsive pacing, and multimodal support. Importantly, even when assistive technologies improve task performance, increases in CL or physiological load may limit usability, scalability, and sustainable real-world deployment in aging populations. By demonstrating the value of multimodal CL assessment and tailored support strategies, this study provides actionable evidence to guide the design, evaluation, and deployment of assistive technologies that genuinely promote autonomy and quality of life in the context of active aging. Acknowledgments The authors would like to acknowledge all the participants who voluntarily took part in the study. This work was supported by the Engineering and Physical Sciences Research Council and National Institute for Health and Care Research funds (grant number EP/W031809/1, IMACTIVE). The data presented in this study are available at [35]. Conceptualization: DM, ADN, JH, and SCMethodology: SV and DMSoftware: SVFormal analysis: RVData curation: SVWriting - original draft preparation: SV, RV, and NCWriting - review and editing: JH, SC, ADN, and DMSupervision: DM and SCFunding acquisition: DM and ADN Conceptualization: DM, ADN, JH, and SC Methodology: SV and DM Software: SV Formal analysis: RV Data curation: SV Writing - original draft preparation: SV, RV, and NC Writing - review and editing: JH, SC, ADN, and DM Supervision: DM and SC Funding acquisition: DM and ADN None declared. None declared. Edited by Matthew Balcarras; submitted 06.Mar.2026; peer-reviewed by Gabriele Pesimena, Simona Massimino; final revised version received 09.Apr.2026; accepted 09.Apr.2026; published 01.Jun.2026. © Simone Varrasi, Roberto Vagnetti, Nicola Camp, John Hough, Alessandro Di Nuovo, Sabrina Castellano, Daniele Magistro. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 1.Jun.2026. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research (ISSN 1438-8871), is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. Please cite as: Varrasi S, Vagnetti R, Camp N, Hough J, Di Nuovo A, Castellano S, Magistro D Human and Robot Assistance for Cognitive Load in Younger and Older Adults: Multimodal Within-Subject Experimental Study J Med Internet Res 2026;28:e94738 doi: 10.2196/94738 PMID: 42224389 Journal of Medical Internet Research ISSN 1438-8871 Copyright © 2026 JMIR Publications

The 'Neo' robot deliberately falls short of realism. (Handout) The 'Neo' robot deliberately falls short of realism. (Handout) At Cooper’s Brewery in Adelaide, the largest family-owned brewing business in Australia, forklifts glide around the sizable factory floor. With ease, they shelve boxes of lagers and ales. But nobody is at the wheel. In fact, there isn’t a wheel at all. Or a seat. These forklifts are automated guided vehicles (AGVs), a pretty old form of robotics that uses lasers, floor markers or other navigation systems to follow pre-planned paths. People had always imagined that robots would look like us, but these forklifts hint at a different future — one in which robots are made in the image of the work, not the worker. For the question of whether robots will be the same shape as humans is not a settled one. In fact, it is a point of some debate in the robotics sector. Elon Musk promises legions of humanoid Tesla robots, poised to take over manual labor and build a colony on Mars. And his is not a particularly wacky view; other executives, including those less given to overconfident predictions, have made similar forecasts. In his lecture delivered to the Cosmos Institute last month in Oxford, the Anthropic co-founder Jack Clark predicted that, by April 2028, bipedal robots will “begin to do useful work in the real world in partnership with human tradespeople”. While being bipedal does not mean a robot is identical to a human, it certainly denotes it being closer in form to us than to a forklift. As recently as this time last year, a wide rollout of electronic colleagues looked unlikely. Even in China, the world leader of the humanoid robotics market, the stumbling and malfunctioning first cohort of robot competitors in the Beijing half-marathon were treated as more of an entertaining sideshow than a serious sign of things to come. Clark himself appeared relatively pessimistic about the capabilities of humanoid robots at the time; in response to an Amazon research paper about the performance of a specialized robotics system for stowing items in the company’s warehouses, he noted that state-of-the art automation was coming from designing highly specific hardware and “carefully structuring your system around a few core tasks”. The experiment, he wrote, “should temper our expectations for bipedal robots leading to some massive improvement in automation — at least in the short term.” What changed? It may simply be that this massive improvement has arrived. At this year’s Beijing half-marathon, several robots outpaced human participants, even breaking records. This feat had serious implications, demonstrating the kind of durability that might soon make the technology suitable for industrial applications. On that basis, Clark’s prediction — that more human-like machines will start trickling into the workforce next year — starts to sound plausible, though we can expect humans would still be on-site to correct for their defects. Manual labor will be slowly automated. Or it may be that Clark believes the underlying AI systems are becoming so powerful, so fast, that the obstacles will be swept away. The robots’ cognitive abilities will catch up with their strength and dexterity. Manual labor will be rapidly automated. “Truly general intelligence would obviate some of the flaws,” Clark wrote in his sceptical 2025 missive. “So if bipeds arrive at the same time as a generally capable intelligence, I’ll need to eat my words.” This is now the precipice at which we find ourselves. As AI improves, those in the field say robots will become more capable of learning for themselves, of adapting to new environments without pre-programming, and even employing reasoning. They’ll be easily able to work in an Amazon warehouse — and then some. The ultimate prize, as Clark says, would be truly general intelligence: a system that can turn its hand to any task without specific training. Many imagine that, in this scenario, the machine most suited to carrying around a general, human-like intelligence would be in the shape of a person. After all, the world is built for humans, and humans have the dexterity to adapt to a vast range of manual tasks. Many industrial robots, by contrast, are made to complete one specific task from a set point. Or else they operate entirely inside facilities built for them: look, for example, at the fulfillment centers of British grocery delivery giant Ocado, where a “hive” of robots roll across a grid system, picking and packing orders. The question is whether the economy will demand robots that can venture into unexpected environments. For this, humans are far more adaptable, taking on stairs and uneven terrain. We climb ladders and traverse scaffolding and even rooftops. It’s hard to imagine a machine able to do all that if it isn’t shaped like us. Clark is probably correct that the bipedal assistants will arrive sooner rather than later, precisely because they might be a stopgap between human work and a totally different way of making things. The trades of the future could look more like the work of British startup Automated Architecture, which makes mobile, robotic micro-factories that can build pre-fabricated components of a house quickly and closer to its final site. Inside the micro-factory, one large robotic arm performs tasks that might typically require several different production lines, producing floors, walls and roofs. Human workforces are still needed to assemble the timber-frame panels, but the process minimizes how much construction is needed on-site by pre-making all the elements for the shell of a building. Think of it like buying a gingerbread house kit instead of baking all the pieces from scratch. In this way, the robotics of industry can go beyond the human approach to a task. Why climb a ladder when you could send a drone? These are already used for surveys and roof inspections. With the right dexterity, perhaps they will be able to perform the fixes as well. Humanoids, argues Josef Chen, the founder of London-based restaurant industry robotics firm Kaikaku AI, are like the equivalent of a pick-up truck on a construction site. “Even though it’s not used for any serious work, people just have it for optionality.” In this analogy, you turn to a digger for digging, a crane for lifting. You wouldn’t expect the truck to do those specialist tasks, even though it can be useful for transportation, towing, or navigating tricky ground. Those who believe in the utility of the human-shaped robot, though, believe that getting it right could unlock huge gains — and huge threats to the human workforce. As Clark wrote back in 2023, “true economic growth from AI happens when you don’t need to design for robots”. If an army of person-shaped droids start arriving in human workplaces, with no adjustments needed for them, it is easy to envision how this would accelerate workforce displacement. Matvey Boguslavskiy, a hardware researcher and director of the Society of Technological Advancement, doesn’t buy that. “I think the future economy is going to have robots with many different form factors, doing many different things,” he says. Indeed, if robot arms or vehicles or dog-like quadrupeds prove vastly more efficient than humans, the argument about whether the world has to be redesigned for robots falls flat: the economic incentive will be enough to justify a redesign. By Alys Key So if applications in the workplace favor the specialized machinery end of the robotics scale, then perhaps the promise of the humanoid robot is in the home. Whereas a factory might have space for several machines, each built for a different purpose, homes are more cramped. It would be difficult and expensive to accommodate an ironing robot, a cooking robot, a gardener robot, and so on — but what if they were all rolled into one? . But here the robots will encounter another obstacle: human disquiet. There is something freakish about Tesla’s Optimus, or Boston Dynamics’ Atlas. The way they are lined up like an army in some promotional videos doesn’t help. It’s also in how close they are to being human-shaped, yet how far, almost like there’s a real person under there wearing a creepy costume. It’s the almost-ness that gets you. The heads are almost the right size, the hands almost lifelike — if often clothed in serial killer black gloves. It may be that we’re too conditioned by science fiction horrors to accept humanoid robots. But the reason those fictional visions terrify us in the first place is because they tap into a primal fear — of a monster, or döppelganger. To make robotics more palatable in the home, many companies are making devices less threatening by ramping up their apparent cuteness. Pixar has a lot to answer for here. Everyone I speak to on this subject cites Wall-E and the success the animation studio had in making a boxy, metal robot into a lovable character. One startup robotics lab even brought in Toy Story screenwriter Alec Sokolow as creative director when developing Ongo, a smart desk lamp with big, Disney-worthy eyes. There seems to be strong consumer demand for such devices, at least among gadget obsessives. There are countless Kickstarters for small robotic companions and widgets that receive hundreds of thousands of dollars in advance orders. However, an interactive paperweight can only do so much. To load the dishwasher, you need grip, spatial reasoning, and a way to get high enough to open or close it. If we really want a smoother life with no chores, there would seem to be few options beyond either hiring another human to help or letting humanoids into the home. What’s the way forward? Boguslaviskiy thinks one of the most useful forms will be the small, wheeled vehicle. We have already seen these catch on for vacuuming and mowing lawns autonomously. Why not for other tasks too? Or if the human shape proves useful, the most acceptable forms might be some compromise between the humanoid and the cute. Take a look at Fauna Robotics, a New York-based company which was acquired by Amazon in March. Their products are deliberately fun to look at, with bright colors and jaunty eyebrows. Crucially, they avoid uncanny valley reactions by not trying to make the robot — named Sprout — look too human. It has Lego-like hands and a head that resembles a Wi-Fi modem. It looks like what it is: a machine. Sprout is also small, at 107cm tall, so won’t tower menacingly over a child. Or there’s Memo, from California-based Sunday Robotics. Sometimes described as humanoid, the robot nevertheless does away with cumbersome feet, instead rolling around on wheels. With its permanent baseball cap, it looks more like a character in a children’s film than a chilling centurion of the future. These non-threatening interfaces serve a deeper purpose than just saving children a fright. They strive to make it possible for us to live alongside intelligent machines. Though AI continues to seep into life and work, there are still many people who avoid using the technology or even encountering it as much as possible. When AI enters the physical world, it becomes less easy to ignore, and triggers in some people a sense of revulsion and anger. We have seen this in the repeated attacks on delivery robots and driverless cars. People in the tech industry who delight in the miracle of creating motion, who have fun hacking together hardware projects on the side, perhaps underestimate how far there is to go on public acceptance of robots. Creepy, faceless, six-foot Cybermen are probably not going to help. Alys Key is a freelance journalist who covers technology, business and policy. She writes the UK 2.0 newsletter on Substack. Δ Δ We welcome applications to contribute to UnHerd – please fill out the form below including examples of your previously published work. Please click here to submit your pitch. Please click here to view our media pack for more information on advertising and partnership opportunities with UnHerd.

BOOM Library has released MECHS & ROBOTS, which they describe as a professional sound effects library built to solve a problem every mech sound designer knows: starting from scratch on every project because nothing in your library was built to work together at scale. Here's the details direct from the company... Eight fully designed robot characters across four size classes share a single design language, so a Microbot and a Titan can coexist in the same session and occupy the same sonic space cleanly. The Construction Kit provides the source material to extend any character or build new ones, giving sound designers a complete mechanical design system in a single library. MECHS & ROBOTS is available in three editions: Designed, Construction Kit, and Bundle. About The Library The Designed edition contains eight fully realised robot characters: Microbot and Nanotron (Small), Android, Automaton, and Summoner (Medium), Gunbuster, Colossus, and Titan (Large), plus Extra Large characters. Each includes 12 sound categories and up to 57 individual variations. Sound categories include Footstep, Movement, Bodyfall, Damage Impact, Transform, Dismantle, Calculate, Expression, Rotation, Power Up, Power Down, and Idle sequence. Each character includes Footstep (10 variations) and Movement (10 variations) files. The Construction Kit is organised into four component categories: Metal, Mechanical, Servo, and Synthetic. Metal recordings cover footsteps, movements, impacts, scrapes, rattles, and debris at small, medium, and large sizes. Mechanical recordings include gears, clunks, clicks, levers, slides, and rotations sourced from industrial and vintage mechanical objects. The Servo category spans six motor configurations: plastic and metal housing in small, medium, and large, captured from real motors, including 3D printer arms, servo-driven industrial equipment, and robot components. The Synthetic category provides granular and wavetable-processed elements, including braams, glitches, textures, power-up sweeps, telemetry sequences, and vocal-quality droid sounds. All files are delivered at 96kHz / 24-bit WAV with full UCS-compliant metadata embedded for use with Soundminer, Basehead, and Soundly. Axel Rohrbach, Creative Director of BOOM Library, said: "Every mech project we've seen starts the same way: designers sourcing and patching together elements that were never built to work at different scales. MECHS & ROBOTS fixes that. One coherent system, from nanobots to ship-sized mechs, with the designed characters and the raw material to build your own." Key Specifications: Pricing and Availability: MECHS & ROBOTS is available now. Regular Pricing – Designed: $139/119€ / Construction Kit: $209/179€ / Bundle: $265/229€ Introductory 20% Off for 2 weeks – Designed: $111.20/€95.20 / Construction Kit: $167.20/€143.20 / Bundle: $212.00/€183.20 Promo pricing ends June 11th, 2026 More information: Andy Mac shows us around Chat DSP M4L and more 2600 & 101 had a baby Port names, internal routing and more Works of Art for Your Eurorack Modules Deep into retro hardware Copyright Sonic State Ltd © 1995-2026. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission from Sonic State is prohibited.

By commenting, you agree to the Prohibited Content Policy By commenting, you agree to the Prohibited Content Policy News See whats happening in Marketing & Advertising right now Exclusive Read and get insights from specially curated unique stories from editorial Leaders Speak Business leaders sharing their insights Events Explore and discuss challenges & trends in India's leading B2B events Awards Recognise work that not only stood out but was also purposeful Webinars Join leaders & experts for roundtables, conferences, panels and discussions Subscribe to our Daily Newsletter By continuing you agree to our Privacy Policy & Terms & Conditions Advertise With Us We have various options to advertise with us including Events, Advertorials, Banners, Mailers, etc. Download ETBrandEquity App Save your favourite articles with seamless reading experience Get updates on your preferred social platform Follow us for the latest news, insider access to events and more. About Us Contact Us Newsletters

Seoul : There were no naked robots in sight at a fashion show held in Seoul with a high-tech twist, where pairs of people and humanoids hit the catwalk in matching outfits. A tasselled blue Texan-style ensemble -- complete with a cowboy hat for the robot -- and a retro silver puffer jacket were among the looks showcased at the event on Thursday. Each human model and their shorter android companion took turns to strut their stuff in unison on stage. The designs, including silky dresses and billowing space-age black trousers like those worn by rock star David Bowie in the 1970s, were carefully fitted to the robots' skeletal frames. Galaxy Corporation, the entertainment company behind the display, said it was meant to ask: "How can humans and robots coexist?" "We realised that robots, too, need to wear clothes," CEO Choi Yong-ho said. "Just as every human being is unique, we believe that every single robot should also be distinct." The clothes were designed by the company, whose spokesperson said it hopes to launch them under the brand name "MACH 33" at the end of the year. The robot models at the Seoul fashion show appeared to be a humanoid made by Chinese startup Unitree, which are popular due to their relatively low cost. Increasingly dexterous robots have proven themselves capable of performing choreographed dances, participating in races, and even able to land backflips. Financial services firm Morgan Stanley predicts the world could have more than a billion humanoids by 2050. But fully automated robots -- using emerging physical AI technology -- are still rare, with most impressive displays remotely operated or pre-programmed.

NEW YORK, May 31. /TASS/. US-based Foundation Future Industries related to the family of President Donald Trump held trials of android robots in the combat operations zone in Ukraine, CNBC television said, citing the company CEO. According to CNBC, the company sent two Phantom MK-1 model robots for the pilot demonstration. This was the first case of using robots in the combat zone. Their core function was to deliver ammunition to the frontline, company CEO Sankaet Pathak said. The model has a number of weaknesses, the TV channel said. Its carrying capacity is not above 20 kg, and the body is not water-proof. The short time of robots’ independent operation became one more obstacle to their large-scale use. The company plans now to send an improved version to Ukraine. The Phantom 2 will boast "superman" abilities, and its carrying capacity will be twice above the one of the Phantom MK-1.

Personalise the news and stay in the know Emergency Backstory Newsletters 中文新闻 BERITA BAHASA INDONESIA TOK PISIN Find any issues using dark mode? Please let us know By Meagan Dillon Topic:Biotechnology Industry Mon 13 Mar 2023 at 6:17am Scientists at a South Australian university are using biology from insects to build robots with a brain – technology that could become a game changer for police, defence and national security. "I'm giving a robot a brain so it can understand its environment," said Flinders University associate professor for autonomous systems, Dr Russell Brinkworth. His biologically-inspired robots have the ability to not just take a picture of the world, but interpret the surrounding environment and adapt accordingly. "Our current robots work well in structured environments that don't change. That sounds complex – but they're all the same," Dr Brinkworth said. "When cars are set on an assembly line, all the cars are the same, the parts are the same, they're all in the same environment, same lighting, same size. "But real life isn't like that – there's always changes, so we need to build robots that can adapt to the environment, rather than forcing the environment to adapt to the robots." Dr Russell Brinkworth studied the biology of insects to create a robot that can understand the vision it captures from a camera. (ABC News: Michael Clements) By studying the biology of insects, such as flies and dragonflies, Dr Brinkworth "reverse engineered" the process that goes on biologically to create a robot that can understand the vision it is capturing from a camera. "The way that insects interpret the world is very similar to the way primates and even humans interpret the world. We just do it on a larger scale," he said. He said his robots can pinpoint a drone or surveillance balloon from kilometres away, far advancing the technology of current cameras. "If you look at security footage and there's someone wearing a hoodie and they've got their face covered – it's hard to understand what's going on based on the camera footage," he said. "But if you were standing there, and you were looking at them, that shadow wouldn't fool your eye because your eye is sophisticated enough to be able to extract out the information hiding in the shadows. "That's the sort of cameras we're building." The robot can understand the vision it captures from a camera, far advancing the technology of current cameras. (:ABC News) Dr Brinkworth said the cameras would be vital for law enforcement, defence and national security in the future, but could also be used for conservation, such as recording the density of koalas in a national park. "You'd have to send somebody out to walk through that forest to do a manual count, or you send a drone overhead with an operator," he said. "They will miss quite a few. "What we've been able to do is design a system that is far more accurate and doesn't rely on human judgement." He said his camera can more accurately pick out where the koalas are because it could "break the different types of camouflage" created by the dense branches and leaves. "It can find animals hiding in forests, it can find all sorts of things that are camouflaged to regular camera views because it's able to look for, and enhance, the very subtle differences across different wavelengths of light." Dr Malcolm Davis says we need robots to be able to understand their environment and make their own decisions. (ABC News: Nick Haggarty) Australian Strategic Policy Institute senior analyst Dr Malcolm Davis said studying biology was important to advance technology such as robots. "We want them to be flexible and in effect, to be able to operate as humans do," he said. "The battle space is a complex, rapidly changing environment – we can't afford to have autonomous systems or robots, essentially requiring human oversight. "We need them to be able to understand their environment and make their own decisions. "There's obviously a Terminator mindset here that if we give robots too much control, then they turn on us [but] I don't think that's likely. "If we're talking about artificial intelligence, if we're talking about advanced robotic systems, then it's humans that are creating those systems. "We would build into those machines an understanding or basis for behaviour based around our approaches to things like international law, humanitarian law, the laws of armed conflict." Dr Brinkworth says the future applications of the robot are in science fiction territory. (CBS Interactive) Dr Brinkworth said the technology could one day be used to potentially build an eye for someone who was blind. "If somebody was blind, you can't just put a camera on them, you have to find a way of giving that information to the person," he said. He said in order to do that, the way the brain communicates needs to be replicated by better understanding how the visual systems of animals work. "We could augment existing visual systems and replace missing parts and communicate that information in the brain's native language back to the brain," he said. "This is very science fiction – it doesn't exist just yet. The interface between technology and biology is a frontier. "The augmentation side of technology moving into biology is a bit Borg-like, if we're looking for Star Trek. "It is very far in the future. But the future starts now, you must start somewhere. "If you want to invest in a future where giving sight back to the blind is important, than you've got to start." Mon 13 Mar 2023 at 6:17am Topic:One Nation Topic:Defence Industry Topic:Unrest, Conflict and War Topic:Electric Vehicles Topic:Weather Topic:Defence Industry Topic:One Nation Topic:Explainer Adelaide Bedford Park Biotechnology Industry Robotics SA Scientific Research Topic:One Nation Topic:Defence Industry Topic:Unrest, Conflict and War Topic:Electric Vehicles Topic:Weather Topic:One Nation Sun 31 May 2026 at 9:14pm Topic:Unrest, Conflict and War Sun 31 May 2026 at 11:35pm Topic:Animals Sun 31 May 2026 at 6:41pm Topic:Olympic Games Venues Sun 31 May 2026 at 4:21pm Your home of Australian stories, conversations and events that shape our nation. This service may include material from Agence France-Presse (AFP), APTN, Reuters, AAP, CNN and the BBC World Service which is copyright and cannot be reproduced. We acknowledge Aboriginal and Torres Strait Islander peoples as the First Australians and Traditional Custodians of the lands where we live, learn, and work. Sign up to get the latest on your favourite topics from the ABC