基于可拓决策和人工势场法的车道偏离辅助系统研究

Study on Extension Decision and Artificial Potential Field Based Lane Departure Assistance System

为保障驾驶安全,车道偏离辅助系统成为一大研究热点。针对车道偏离辅助系统中的决策问题,提出采用可拓决策方法:以跨道时间和距车道中心线距离作特征量建立二维可拓集合,设计动态边界划分可拓域,不同域中分别采用自由驾驶模式、辅助驾驶模式和线控主动转向模式;求解出关联函数,确定可拓域中辅助驾驶模式人和车的控制权重。针对车道偏离辅助系统的横向运动控制问题,采用人工势场法设计转角控制器:建立了跨道时间模型,以跨道时间为变量设计势场函数,以得到的势场力代入参考车辆模型,用PID控制跟踪理想转向盘转角。针对不同域表征的车道偏离状态,实现自由驾驶模式、辅助驾驶模式和线控主动转向模式之间的切换。利用Carsim/Simulink联合仿真和硬件在环试验台验证了该方法的有效性。

In order to ensure the driving safety, the research on the lane departure assistance system becomes hot recently. For the decision-making problem in the lane departure assistance system, the extension decision method is put forward: the time to lane crossing(TLC) and the distance to lane center(DLC) are used as the characteristic variables to establish the two-dimensional extension set, the dynamic boundary is designed to divide extensive domains, and the free driving mode, assistant driving mode and active steering-by-wire mode are adopted separately in the different domains; the correlative function is calculated to determine the control weights of human and vehicle of the assistant driving mode in the extensive domain. For the lateral motion control of the lane departure assistance system, the artificial potential field method is used to design the steering angle controller: the TLC model is established, the potential function is designed with TLC, the obtained potential force is substituted into the vehicle reference model, and PID control is used to track the ideal steering angles. For the lane departure state represented in different domains, the switching is achieved among the free driving mode, assistant driving mode and active steering-by-wire mode. The validity of the proposed method is verified through the Carsim/Simulink joint simulation and hardware-in-loop test.