Video Friday this is your weekly selection of amazing robotics videos collected by your friends on IEEE spectrum robotics. We're also posting a weekly calendar of upcoming robotics events for the next few months. Please send us your events to turn on.
ROSCON 2025: October 27–29, 2025, SINGAPORE
Enjoy today's videos!
Welcome to this world! Height 180 cm, weight 70 kg. A bionic humanoid H2, born to serve everyone safely and kindly.
From $29,900 plus tax and shipping.
[ Unitree ]
The title of this article, “An Eagle Stole Our FPV Drone,” pretty much sums it up.
[ Team BlackSheep ]
Thank you, Kyu-Jin!
FYI: Any robot that sounds like a delicious baked good is guaranteed favorable coverage on Video Friday.
[ Cleo Robotics ]
Oli now performs a smooth and coordinated sequence of full-body movements from lying down to rising. Oli, 165 cm tall and with 31 degrees of freedom, continues to demonstrate natural and fluid movements.
[ LimX Dynamics ]
Thank you, Jinyan!
Blog Friend Bram Vanderbort inspects the exhibition hall in I'll go 2025 in Hangzhou, China.
[ IROS 2025 ]
In a fireside chat with Professor Sam Madden, Ty Brady, Chief Technology Officer Amazon Roboticsdiscuss the trajectory of robotics development and how generative AI plays a role in robotics innovation.
[ MIT Generative AI Impact Consortium ]
Professor Dimitrios Kanoulas gave an invited talk at the workshop “The Art of Reliability: Overcoming Failures in Robotics” at IROS 2025.
[ IROS 2025 ]
The GRASP talk at the University of Pennsylvania was given by Suraj Nair from the Department of Physical Intelligence on the topic “Scaling Robot training with Vision-Language-Action models.
Over the past few years, we have seen enormous advances in the capabilities of artificial intelligence systems, driven in large part by foundation models that scale expressive architectures with different data sources. While the impact of this technology on vision and language understanding is clear, its use in robotics is still in its infancy. Scaling robot learning still presents numerous open challenges, from choosing the right data to scale to design. algorithms which can effectively use this data to work with feedback in the physical world. At Physical Intelligence, we strive to address these questions. This talk will present our recent work on building vision-language-action models, covering topics such as architectural design, data scaling, and open research directions.
[ University of Pennsylvania GRASP Laboratory ]
Articles from your site
Related articles on the Internet
