Robust Locomotion of Legged Robots with Closed-Loop Guarantees
This talk presents a predictive control framework allowing general legged robots to walk robustly, and describes its validation results on both humanoids and quadrupeds.
Overview
The development and analysis of a controller enabling the locomotion of a general legged robot (e.g., biped or quadruped) is presented. The controller, modifying a nonlinear Model Predictive Control - MPC - scheme, ensures closed loop stability and robustness to a class of constant disturbances. This is achieved thanks to employing the machinery of adaptive control together with Control Lyapunov Functions. Experimental validation of the controller on both a humanoid robot (the Italian ergoCub) and a quadruped (the unitree Aliengo) are presented. This work is now part of open source software packages published by the Italian Institute of Technology.
Presenters
Brief Biography
Mohamed Elobaid is a Research Scientist at King Abdullah University of Science and Technology (KAUST). Prior to joining KAUST, he was a Postdoctoral researcher with the Italian Institute of Technology in Genoa, where he was responsible for a joint research lab with Honda R&D. Elobaid obtained his PhD in Automatic Control (magna cum laude) from Sapienza University of Rome in cotutelle with University of Paris Saclay (2022). Elobaid works on control and optimization for robotic and mechatronic systems, spanning both theory and deployment.
Related People
Related Researchers
Mohamed Elobaid
- Research Scientist, Electrical and Computer Engineering