Robust Predictive Control for Quadrupedal Locomotion: Learning to Close the Gap between Reduced- and Full-Order Models

• Published in: IEEE robotics and automation letters Vol. 7; no. 3; p. 1
• Main Authors: Pandala, Abhishek, Fawcett, Randall, Rosolia, Ugo, Ames, Aaron, Akbari Hamed, Kaveh
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.07.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Computational modeling; Controllers; Deep learning; Legged locomotion; Legged Robots; Locomotion; Mathematical models; Motion Control; Multi-Contact Whole-Body Motion Planning and Control; Neural networks; Nonlinear control; Planning; Predictive control; Quadrupedal robots; Real-time systems; Reduced order models; Reduced order systems; Rigid structures; Robot dynamics; Robust control; Trajectory optimization; Trajectory planning; Uncertainty
• ISSN: 2377-3766; 2377-3766
• DOI: 10.1109/LRA.2022.3176105

Template-based reduced-order models have provided a popular methodology for real-time trajectory planning of dynamic quadrupedal locomotion. However, the abstraction and unmodeled dynamics in template models significantly increase the gap between reduced- and full-order models. This letter presents a computationally tractable robust model predictive control (RMPC) formulation, based on convex quadratic programs (QP), to bridge this gap. The RMPC framework considers the single rigid body model subject to a set of unmodeled dynamics and plans for the optimal reduced-order trajectory and ground reaction forces (GRFs). The generated optimal GRFs of the high-level RMPC are then mapped to the full-order model using a low-level nonlinear controller based on virtual constraints and QP. The proposed hierarchical control framework is employed for locomotion over rough terrains. We leverage deep reinforcement learning to train a neural network to compute the set of unmodeled dynamics for the RMPC framework. The proposed controller is finally validated via extensive numerical simulations and experiments for robust and blind locomotion of the A1 quadrupedal robot on different terrains.