A quantitative evaluation about the safety of intended functionality (SOTIF) for adaptive cruise control based on extension optimization

Regarding the safety of the intended functionality (SOTIF) of the existing adaptive cruise control (ACC) algorithm in the context of curved roads, this paper proposes a novel adaptive cruise control algorithm. This algorithm is predicated on the extension of coordinated control optimization. The study aims to address the insufficient consideration of lateral and longitudinal dynamics coupling effects by conventional algorithms under complex curved-road driving conditions, as well as the absence of a systematic quantitative evaluation framework for SOTIF. The objective of this proposal is to enhance the functionality of the adaptive cruise control system in scenarios involving curves and variations in road adhesion coefficients. To this end, a quantitative evaluation system has been developed to analyze the safety risk scores of control algorithms under these scenarios. Initially, an analysis of the adaptive cruise system is conducted to ensure the SOTIF demand. This analysis employs a system theoretic process analysis (STPA)method and proposes a functional enhancement goal for the curved road scenario. Secondly, a control strategy for prediction based on lateral and longitudinal coupling is constructed using extension coordinated control theory. A quantitative evaluation system is constructed based on the national standard through the joint simulation of Carsim and Matlab/Simulink. The efficacy of the enhanced algorithm is substantiated by the observation that it successfully reduces the traditional algorithm's score.