The ACC safety research and application based on system theory process analysis

Abstract

With the development of autonomous driving technology, more and more attention is paid to the safety of autonomous driving system, such as functional safety, safety of intended functionality and cyber security. Among them, the safety of intended functionality refers to the safety problems caused by the insufficient function of automatic driving system and the harm caused by the misuse of personnel. System theory Analysis method (System Theoretic Process Analysis, STPA) is a kind of analysis method which treat the whole system as one object. In this paper, STPA analysis method is applied to adaptive cruise control (ACC) system, the control schematic of ACC system is constructed, the unsafe control behavior of the system is analyzed, and the system constraints and typical test scenarios are obtained. Taking the front vehicle cutting in scenario as an example, through using autonomous robots and navigable target vehicle, the obstacle avoidance ability of the driver for the front vehicle cutting in scenario with different speeds at 120km/h is tested. When the preceding vehicle cutting in speed is 60km/h, the collision can be avoided whereas the cutting distance is 31.8m or above. When the preceding vehicle cuts in at the speed of 80km/h, the collision can be avoided whereas the cutting distance is 16.7m or above. This index can be used as the requirements for ACC system in the cut-in scenario. In this paper, the ACC system is analyzed in detail for the safety of intended functionality and relevant requirements are given. The whole method and steps are also applicable to other automatic driving systems.