Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems

Abstract

Many modern nonlinear control methods aim to endow systems with guaranteed properties, such as stability or safety, and have been successfully applied to the domain of robotics. However, model uncertainty remains a persistent challenge, weakening theoretical guarantees and causing implementation failures on physical systems. This paper develops a machine learning framework centered around Control Lyapunov Functions (CLFs) to adapt to parametric uncertainty and unmodeled dynamics in general robotic systems. Our proposed method proceeds by iteratively updating estimates of Lyapunov function derivatives and improving controllers, ultimately yielding a stabilizing quadratic program model-based controller. We validate our approach on a planar Segway simulation, demonstrating substantial performance improvements by iteratively refining on a base model-free controller.

References in corpus (6)

• Control Barrier Function Based Quadratic Programs for Safety Critical Systems
• Neural Lander: Stable Drone Landing Control using Learned Dynamics
• Safe Controller Optimization for Quadrotors with Gaussian Processes
• Stable Gaussian Process based Tracking Control of Euler-Lagrange Systems
• Safe Learning of Regions of Attraction for Uncertain, Nonlinear Systems with Gaussian Processes
• Learning Lyapunov (Potential) Functions from Counterexamples and Demonstrations

Cited by in corpus (8)

• Adaptive CLF-MPC With Application To Quadrupedal Robots
• A Control Barrier Perspective on Episodic Learning via Projection-to-State Safety
• A Control Lyapunov Perspective on Episodic Learning via Projection to State Stability
• Neural Identification for Control
• Statistical Guarantees in Data-Driven Nonlinear Control: Conformal Robustness for Stability and Safety
• Model-Free Reinforcement Learning for Symbolic Automata-encoded Objectives
• Continuous Lyapunov Controller and Chaotic Non-linear System Optimization using Deep Machine Learning
• Learning-based synthesis of robust linear time-invariant controllers