Quantifying Safety of Learning-based Self-Driving Control Using Almost-Barrier Functions

TL;DR

This paper introduces a method to quantify the safety of deep learning-based self-driving controllers by learning almost-barrier functions that approximate safe operating regions, enabling safety analysis and online monitoring.

Contribution

It presents a novel learning and certification framework for almost-barrier functions to analyze and ensure the safety of neural controllers in autonomous driving.

Findings

Effective safety quantification in simulation environments

Applicable to simple kinematic and complex vehicle dynamics

Enables online safety monitoring through reachability analysis

Abstract

Path-tracking control of self-driving vehicles can benefit from deep learning for tackling longstanding challenges such as nonlinearity and uncertainty. However, deep neural controllers lack safety guarantees, restricting their practical use. We propose a new approach of learning almost-barrier functions, which approximately characterizes the forward invariant set for the system under neural controllers, to quantitatively analyze the safety of deep neural controllers for path-tracking. We design sampling-based learning procedures for constructing candidate neural barrier functions, and certification procedures that utilize robustness analysis for neural networks to identify regions where the barrier conditions are fully satisfied. We use an adversarial training loop between learning and certification to optimize the almost-barrier functions. The learned barrier can also be used to construct online safety monitors through reachability analysis. We demonstrate effectiveness of our methods in quantifying safety of neural controllers in various simulation environments, ranging from simple kinematic models to the TORCS simulator with high-fidelity vehicle dynamics simulation.