AMBER Lab

I am so happy to share that our paper entitled ``Distributed data-driven predictive control for multi-agent collaborative legged locomotion" by my Ph.D. students R. Fawcett, L. Amanzadeh, and J. Kim, in collaboration with Dr. A. Ames (@AMBER_lab), has been selected as the outstanding paper of the 2023 IEEE International Conference on Robotics and Automation (#ICRA2023). arxiv.org/abs/2211.06917

1) Cooperative locomotion arises in different applications of legged robots, such as cooperative transportation of payloads like social insects. Models of these cooperative locomotion systems are inherently complex; they are nonlinear with hybrid dynamics and high dimensionality.

1) Our recent paper with @AMBER_lab presents a "data-driven" model for real-time trajectory planning of dynamic quadrupedal robots. Many of the existing approaches for planning or controlling legged robots rely on the robot's "physics-based" reduced-order (i.e., template) models.

1) In recent work with @AMBER_lab, we have presented a layered control approach, based on robust model predictive control (RMPC), nonlinear control, and reinforcement learning (RL), to learn to close the gap between reduced- and full-order models of quadrupedal locomotion.