Boston Dynamics and Toyota Research Institute Collaborate on AI-Powered Humanoid Robot

Boston Dynamics and Toyota Research Institute have demonstrated a Large Behavior Model (LBM) powering the Atlas humanoid robot. The LBM enables the robot to perform autonomous whole-body manipulation and locomotion behaviors. The model allows for new capabilities to be added quickly without hand-programming, and the robot can self-adjust to unexpected physical challenges. This breakthrough is the result of a joint research partnership and reaffirms the potential of AI technologies in developing general-purpose humanoid assistants.

Boston Dynamics and Toyota Research Institute (TRI) have achieved a significant milestone in the field of robotics and artificial intelligence with the demonstration of a Large Behavior Model (LBM) powering the Atlas humanoid robot. This breakthrough, announced on August 20, 2025, showcases the robot's ability to perform autonomous whole-body manipulation and locomotion behaviors [1].

The LBM enables Atlas to accomplish complex tasks that require combining object manipulation with locomotion. The robot can perform a series of packing, sorting, and organizing tasks, demonstrating its dexterity and adaptability. Throughout the sequences, Atlas can self-adjust to unexpected physical challenges, such as closing the lid of a box and sliding it across the floor [1].

The joint research partnership between Boston Dynamics and TRI, established in October 2024, has been instrumental in accelerating the development of smart robots. By leveraging their combined strengths and expertise, the collaboration has resulted in a breakthrough that reaffirms the incredible potential of AI technologies in developing general-purpose humanoid assistants [1].

Scott Kuindersma, vice president of Robotics Research at Boston Dynamics, commented, "Training a single neural network to perform many long-horizon manipulation tasks will lead to better generalization, and highly capable robots like Atlas present the fewest barriers to data collection for tasks requiring whole-body precision, dexterity, and strength" [1].

Russ Tedrake, senior vice president of Large Behavior Models at TRI, added, "One of the main value propositions of humanoids is that they can achieve a huge variety of tasks directly in existing environments, but the previous approaches to programming these tasks simply could not scale to meet this challenge. Large Behavior Models address this opportunity in a fundamentally new way – skills are added quickly via demonstrations from humans, and as the LBMs get stronger, they require less and less demonstrations to achieve more and more robust behaviors" [1].

This breakthrough is a significant step forward in the development of general-purpose humanoid robots. The ability to quickly add new capabilities without hand-programming and the robot's capacity to self-adjust to unexpected challenges are key features that could transform how we live and work. As the technology continues to evolve, it will be crucial to monitor both the technological progress and its societal implications.

References:
[1] https://www.prnewswire.com/news-releases/ai-powered-robot-by-boston-dynamics-and-toyota-research-institute-takes-a-key-step-towards-general-purpose-humanoids-302534045.html