SafeBench: A Benchmarking Platform for Safety Evaluation of Autonomous Vehicles

TL;DR

SafeBench is a comprehensive benchmarking platform that unifies various safety-critical testing scenarios and algorithms for autonomous vehicles, enabling fair comparison and fostering development of safer autonomous driving systems.

Contribution

The paper introduces SafeBench, the first unified platform for autonomous vehicle safety testing, integrating diverse scenarios, generation algorithms, and enabling fair evaluation of different AD algorithms.

Findings

Generated scenarios are more challenging for AD agents.

Trade-off observed between performance in benign and safety-critical scenarios.

SafeBench facilitates large-scale, effective testing of autonomous driving algorithms.

Abstract

As shown by recent studies, machine intelligence-enabled systems are vulnerable to test cases resulting from either adversarial manipulation or natural distribution shifts. This has raised great concerns about deploying machine learning algorithms for real-world applications, especially in safety-critical domains such as autonomous driving (AD). On the other hand, traditional AD testing on naturalistic scenarios requires hundreds of millions of driving miles due to the high dimensionality and rareness of the safety-critical scenarios in the real world. As a result, several approaches for autonomous driving evaluation have been explored, which are usually, however, based on different simulation platforms, types of safety-critical scenarios, scenario generation algorithms, and driving route variations. Thus, despite a large amount of effort in autonomous driving testing, it is still challenging to compare and understand the effectiveness and efficiency of different testing scenario generation algorithms and testing mechanisms under similar conditions. In this paper, we aim to provide the first unified platform SafeBench to integrate different types of safety-critical testing scenarios, scenario generation algorithms, and other variations such as driving routes and environments. Meanwhile, we implement 4 deep reinforcement learning-based AD algorithms with 4 types of input (e.g., bird's-eye view, camera) to perform fair comparisons on SafeBench. We find our generated testing scenarios are indeed more challenging and observe the trade-off between the performance of AD agents under benign and safety-critical testing scenarios. We believe our unified platform SafeBench for large-scale and effective autonomous driving testing will motivate the development of new testing scenario generation and safe AD algorithms. SafeBench is available at https://safebench.github.io.