Towards Data-Driven Model-Free Safety-Critical Control

TL;DR

This paper introduces a data-driven framework using neural networks and probabilistic safety conditions to enable safe, model-free velocity control in robotic systems, addressing practical tuning challenges of control barrier functions.

Contribution

It develops a novel approach combining neural network-based Lyapunov function learning with probabilistic safety bounds to improve model-free control barrier functions in robotics.

Findings

Successfully tested on UR5e robot with demonstrated safety improvements

Effectively estimates decay rates for velocity controllers in practice

Provides a robust safety guarantee under uncertainty

Abstract

This paper presents a framework for enabling safe velocity control of general robotic systems using data-driven model-free Control Barrier Functions (CBFs). Model-free CBFs rely on an exponentially stable velocity controller and a design parameter (e.g. alpha in CBFs); this design parameter depends on the exponential decay rate of the controller. However, in practice, the decay rate is often unavailable, making it non-trivial to use model-free CBFs, as it requires manual tuning for alpha. To address this, a Neural Network is used to learn the Lyapunov function from data, and the maximum decay rate of the systems built-in velocity controller is subsequently estimated. Furthermore, to integrate the estimated decay rate with model-free CBFs, we derive a probabilistic safety condition that incorporates a confidence bound on the violation rate of the exponential stability condition, using Chernoff bound. This enhances robustness against uncertainties in stability violations. The proposed framework has been tested on a UR5e robot in multiple experimental settings, and its effectiveness in ensuring safe velocity control with model-free CBFs has been demonstrated.