Learning Local Control Barrier Functions for Hybrid Systems

TL;DR

This paper introduces a learning-based method to develop local Control Barrier Functions for hybrid systems, enabling safe, efficient, and scalable control in complex robotic applications.

Contribution

It proposes a novel learning-enabled approach for constructing local CBFs that guarantees safety in nonlinear hybrid systems, addressing computational and scalability limitations of prior methods.

Findings

The approach is computationally efficient and minimally invasive.

It is applicable to large-scale hybrid systems.

Demonstrated effectiveness in robotic examples including autonomous racing.

Abstract

Hybrid dynamical systems are ubiquitous as practical robotic applications often involve both continuous states and discrete switchings. Safety is a primary concern for hybrid robotic systems. Existing safety-critical control approaches for hybrid systems are either computationally inefficient, detrimental to system performance, or limited to small-scale systems. To amend these drawbacks, in this paper, we propose a learning-enabled approach to construct local Control Barrier Functions (CBFs) to guarantee the safety of a wide class of nonlinear hybrid dynamical systems. The end result is a safe neural CBF-based switching controller. Our approach is computationally efficient, minimally invasive to any reference controller, and applicable to large-scale systems. We empirically evaluate our framework and demonstrate its efficacy and flexibility through two robotic examples including a high-dimensional autonomous racing case, against other CBF-based approaches and model predictive control.