Control Parameter Optimization for Automobile Cruise Control System via Improved Differential Evolution Algorithm

摘要

In the regulation layer, the automobile cruise control system (ACCS) is responsible for transient lateral maneuvers and closely related to executing steady state. This paper develops an improved differential evolution algorithm (IDEA) to deal with the control parameter optimization problem for the ACCS. Based on the classical Bayesian decision strategy, a prior probability based sequential chromosome generator is built to partition the problem space and approach the probable solution domain as close as possible. By mimicking the crossover and mutation behaviors among chromosomes, an online adaptive search method is proposed. The IDEA can effectively deepen the search area and simplify the parameter tuning process to get a well-performed ACCS. A nonlinear automobile model is used as a test bed to verify the feasibility and efficiency of the proposed method. Numerical simulations show that the IDEA optimized ACCS has good performance in terms of both steady state maneuvers and transient maneuvers.