Adaptive cruise control strategy based on fuzzy variable weight model predictive control

Abstract

Aiming at the adaptability and the ability to coordinate different control objectives of the adaptive cruise control (ACC) system in complex urban driving environments, this paper proposes a novel ACC strategy based on model predictive control (MPC) with fuzzy variable weight coefficients of multi-objective optimization. First, according to the requirements of vehicle performance indicators under different driving conditions, the weight coefficients in the MPC framework are updated online by a fuzzy control algorithm to adjust the priorities of safety, economics, and car-following ability. Then, considering that quadratic programming (QP) may fall into a local optimum, the particle swarm optimization (PSO) algorithm is used to solve the optimal control policy online. The effectiveness and reliability of the proposed control strategy are validated on the PreScan/MATLAB/Simulink co-simulation platform and the driver hardware in the loop platform (DHIL) compared to other existing control strategies.