Exploring autonomous vehicle crash risk: system coupling effects and key causal factors.

INTRODUCTION

The continuous progression of autonomous driving technology is propelling the automotive industry into an unprecedented era, with the intelligence and driving safety capabilities of autonomous vehicles serving as crucial benchmarks for assessing industry development. However, crashes involving autonomous vehicles have raised concerns among both government authorities and the general public regarding this technology. Consequently, conducting a comprehensive analysis of crash causes and key causal factors holds immense significance for technological progress, personnel safety, and shaping the future direction of the automotive industry.

METHOD

Based on the AVOID global autonomous vehicle operation event dataset, this study constructs an autonomous vehicle crash causation system and risk coupling metric model, achieved by employing the hiking optimization algorithm, recursive feature elimination method, N-K model, and Bayesian networks to investigate the interplay between risk systems and the coupling effect of autonomous vehicle crashes in complex traffic environments.

RESULTS

(1) The crash causation system of automatic driving constructed in this study has high stability and robustness; (2) The degree of crash risk has a positive correlation with the number of coupled systems, and the important causative factors affecting crash risk include system non-engagement, vehicle depreciation, speeding, abrupt stopping, unrestrained occupants, dark, night and other factors; (3) The role of vehicle defects in the induction of AV crashes should not be ignored, and the poor environmental conditions will significantly enhance the crash risk under the coupling of the three systems.

PRACTICAL APPLICATIONS

This study identifies both the mechanism behind risk system coupling in autonomous vehicle crashes and key causal factors involved therein, offering significant theoretical support for comprehending the underlying causes of AV crashes, thereby aiding in enhancing the safety performance of autonomous vehicle vehicles and informing the development of traffic safety policies in relevant domains.