A Survey on Safety-Critical Driving Scenario Generation—A Methodological Perspective

TLDR

Autonomous driving systems have advanced rapidly, yet their safety evaluation remains difficult because safety‑critical scenarios are rare in real‑world data. This survey focuses on algorithms for artificially generating safety‑critical driving scenarios to assess risk and reduce costs. The authors classify existing generation methods into data‑driven, adversarial, and knowledge‑based categories, review simulation tools, and outline challenges such as fidelity, efficiency, diversity, transferability, and controllability.

Abstract

Autonomous driving systems have witnessed significant development during the past years thanks to the advance in machine learning-enabled sensing and decision-making algorithms. One critical challenge for their massive deployment in the real world is their safety evaluation. Most existing driving systems are still trained and evaluated on naturalistic scenarios collected from daily life or heuristically-generated adversarial ones. However, the large population of cars, in general, leads to an extremely low collision rate, indicating that safety-critical scenarios are rare in the collected real-world data. Thus, methods to artificially generate scenarios become crucial to measure the risk and reduce the cost. In this survey, we focus on the algorithms of safety-critical scenario generation in autonomous driving. We first provide a comprehensive taxonomy of existing algorithms by dividing them into three categories: data-driven generation, adversarial generation, and knowledge-based generation. Then, we discuss useful tools for scenario generation, including simulation platforms and packages. Finally, we extend our discussion to five main challenges of current works– fidelity, efficiency, diversity, transferability, controllability– and research opportunities lighted up by these challenges.