Learning for Safety-Critical Control with Control Barrier Functions

Andrew Taylor,Andrew Singletary,Yisong Yue,Aaron Ames

DOI: https://doi.org/10.48550/arXiv.1912.10099

2019-12-21

Abstract:Modern nonlinear control theory seeks to endow systems with properties of stability and safety, and have been deployed successfully in multiple domains. Despite this success, model uncertainty remains a significant challenge in synthesizing safe controllers, leading to degradation in the properties provided by the controllers. This paper develops a machine learning framework utilizing Control Barrier Functions (CBFs) to reduce model uncertainty as it impact the safe behavior of a system. This approach iteratively collects data and updates a controller, ultimately achieving safe behavior. We validate this method in simulation and experimentally on a Segway platform.

Systems and Control,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to improve the safety of control systems through machine - learning methods in the presence of model uncertainties. Specifically, the authors propose a learning framework using Control Barrier Functions (CBFs) to reduce the impact of model uncertainties on the system's safe behavior. This method ultimately achieves the system's safe behavior by iteratively collecting data and updating the controller. The research particularly focuses on how to ensure the system's safe control under uncertain models, especially in fields such as autonomous driving, industrial robots, and aerospace, where applications have very high safety requirements. The main contribution of the paper is to provide a new method that directly learns how model uncertainties affect the evolution of control barrier functions and synthesizes new controllers accordingly to ensure safety. This method can not only learn the impact of unmodeled dynamics and parameter errors on system behavior but also can be combined with optimization - based safety control methods, thereby ensuring the safety of the system within the framework of nonlinear control theory. In the experimental part, the authors verified the effectiveness of this method on simulation and actual hardware platforms, especially for the application of the Segway platform, demonstrating the ability of the learning - enhanced controller to handle model uncertainties.

Learning for Safety-Critical Control with Control Barrier Functions

Learning Control Barrier Functions with High Relative Degree for Safety-Critical Control

Learning Local Control Barrier Functions for Hybrid Systems

Learning a Better Control Barrier Function Under Uncertain Dynamics

Learning for Layered Safety-Critical Control with Predictive Control Barrier Functions

Reinforcement Learning for Safety-Critical Control under Model Uncertainty, using Control Lyapunov Functions and Control Barrier Functions

Learning Performance-Oriented Control Barrier Functions Under Complex Safety Constraints and Limited Actuation

Recursively Feasible Probabilistic Safe Online Learning with Control Barrier Functions

End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks

Learning Differentiable Safety-Critical Control using Control Barrier Functions for Generalization to Novel Environments

Learning a Formally Verified Control Barrier Function in Stochastic Environment

State-action control barrier functions: Imposing safety on learning-based control with low online computational costs

Safe Nonlinear Control Using Robust Neural Lyapunov-Barrier Functions

Guaranteeing Safety of Learned Perception Modules via Measurement-Robust Control Barrier Functions

On the Optimality, Stability, and Feasibility of Control Barrier Functions: An Adaptive Learning-Based Approach

Safe Online Dynamics Learning with Initially Unknown Models and Infeasible Safety Certificates

Advances in the Theory of Control Barrier Functions: Addressing practical challenges in safe control synthesis for autonomous and robotic systems

Sablas: Learning Safe Control for Black-Box Dynamical Systems

Safe Barrier-Constrained Control of Uncertain Systems via Event-triggered Learning