AthenaZero can juggle barehanded using onboard vision feedback. By using multi-fingered hands, it can transition seamlessly between a wide range of juggling patterns.
Resources
By Topic:
By Type:
Getting a Grip on Robotic Data Collection
To make sure the data collected transfers to the robot, we’ve co-designed handheld and robot grippers: same linkage mechanisms, same degrees of freedom, same force...
Meet “Roadrunner”: a bipedal, wheeled robot for multi-modal locomotion
Roadrunner can stand up from the ground, drive with its wheels side-by-side, or shift them in-line.
A Dynamic Robot That Can Throw, Catch, and Hit a Baseball
Researchers at the RAI Institute have built a low-impedance platform to study dynamic robot manipulation. In this demo, robots play a game of catch and...
From Walking to Working: Spot Stacks Tires
Spot robot performs dynamic whole-body manipulation using a combination of reinforcement learning and sampling-based control. Behavior shown in the video is fully autonomous, including the...
Ultra Mobile Vehicle Expands Mobility Repertoire
Using reinforcement learning we have expanded the range of techniques the Ultra Mobile Vehicle (UMV) uses to handle terrain and obstacles, including hops, out-of-plane balance,...
Balance and Brains: Building an Ultra Mobile Vehicle
While AI has mastered language, getting robots to move swiftly and effortlessly through the unstructured world has not been solved. RAI Institute is building robots...
New Tricks Loading…
Ultra Mobile Vehicle (UMV) learns to flip.
Reinforcement Learning Accelerates Humanoid Behavior Production
Reinforcement learning is used to speed the production of behavior for the Atlas humanoid robot. At the heart of the learning process is a physics-based...
Stunting with Reinforcement Learning
In this demo, Ultra Mobile Vehicle (UMV) drives, turns, jumps, tricks, and comes to a sudden stop called a track-stand. All of the driving, landings, balance, and track-stands are done using reinforcement learning.
Spot Speeds Up
Using reinforcement learning, we trained policies for Spot that allow the robot to achieve record running speeds of 11.5 mph (5.2 m/s) — over three times faster than Spot's default max speed.
Inside the Institute: August 2024 Sneak Peek
A look at the last couple of months, including new dexterity tasks, gait training, bigger jumps, and more.