基于粒子群算法的汽车自适应巡航控制器设计

Design of Vehicle Adaptive Cruise Controller Based on PSO Algorithm

提出了一种基于粒子群优化算法的模糊自校正控制器参数优化方法。基于搭建的Carsim和Simulink联合仿真环境,选取典型优化工况,利用粒子群优化算法对控制器比例因子和隶属度函数形状参数在取值区间内多次随机选值,并重构控制器,发挥算法本身具有的记忆最佳取值点和各点间相互对比机制,以跟踪目标函数为最优,实现对控制器性能优化问题的求解。通过典型工况仿真和实车试验结果表明,该方法优化后的控制器具有更优良的控制性能,可有效降低自适应巡航系统与整车的性能匹配设计工作量,为模糊控制器的参数确定提出了一套可行的研究途径。

Based on particle swarm optimization （PSO） algorithm, a fuzzy self-tuning controller parameters optimization method was developed. With the co-simulation of Carsim and Simulink, typical optimized working conditions were selected. The controller＇ s scaling factor value and the position of membership function shape points were randomly selected to ensure the controller optimal performance. The controller was also reconstructed. Optimum remembering points and contrast mechanism among these points were working in PSO algorithm with the optimum target function. The actual vehicle experiments were carried out under typical working conditions. The experiment results showed that the optimized controller had good control performance, which could decrease the design workload of performance matching between the adaptive cruise control and test vehicle.