Harvard Researchers Develop AI Control System for Legged Robots
A team led by Yilun Du at Harvard's John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Kempner Institute has developed Diffusion-MPC (Model Predictive Control), an AI-based control system for four-legged robots. This system allows robots to quickly adapt their movements to new tasks and terrains, such as changing slopes or balancing, without requiring retraining. Diffusion-MPC integrates a generative diffusion model, similar to those used in image generation, to achieve this flexible adaptation.
This development represents a significant step towards general-purpose locomotion controllers, enabling more versatile and robust robotic applications in varied and unpredictable environments.
Learn one new AI thing every day.
Daily Deck sends you seven plain-English cards like this every morning. Free.
Start free