车辆驾驶员辅助系统中基于运动模型的前车最优跟随控制

Preceding Vehicle Optimal Following Control Based on Kinematics Models for Advanced Driver Assistant System of Motor Vehicles

为解决驾驶员辅助系统在前车跟随控制中车距与车速误差不易同步收敛的问题,建立考虑主车对期望加速度动态响应的主车与前车运动模型,以对前车跟随控制提供模型参考。基于主车实际运动与理想前车跟随行驶间的偏差建立前车跟随误差模型,并用线性二次型调节器求解最优期望加速度控制序列,以作为前车跟随控制中主车运动参数控制的输出。为考虑在实车控制中车辆及环境参数发生变化的可能,在前车跟随误差模型中引入车距与车速误差累积项以使线性二次型调节器具有积分控制功能,以在实际车辆与理想模型发生偏差时仍可进行有效控制。仿真对比及实车试验证明,基于运动模型的前车跟随控制可使主车车距与车速误差同步收敛并有效减少主车达到稳态前车跟随行驶的时间,同时易于在实车控制中展开应用。

For improving the responding characteristics of distance and velocity control in preceding vehicle following scenario of advanced driver assistant system, a kinematics model, been used in the proceeding vehicle following controller, is established based on the interrelated motion of the host and preceding vehicles. Utilizing the error model obtained from desired and actual motion of the host vehicle, optimal desired acceleration, which is the output of host vehicle’s motion controller, is calculated based on the linear quadratic regulator. Accumulation of distance and velocity errors are added into the kinematics model of vehicle following so as to enable the linear quadratic regulator to function with resistance to the changing vehicle or environment parameters when using a standard model. Simulation comparison and vehicle experiments confirm that, while reducing the control time consumption, preceding vehicle optimal following control could improve the responding characteristic of distance and velocity errors and is convenient to be used in the real vehicles.