基于改进势场法的智能车超车动态路径规划

为了改进人工势场法(APF)在智能车高速超车状态下规划出的轨迹规范性差、安全性得不到保障、以及路径不平滑的问题。通过设计用于指引智能车换道的换道引力势场,以保证换道轨迹的规范性；构建异向分布结构的障碍车斥力势场提高换道的安全性；并建立带有转向缺口的道路中心线斥力势场和设计平滑代价函数优化规划出的轨迹点,使规划出的路径更加平滑。基于线性二次型调节器(LQR)设计路径跟踪控制器跟踪规划出的路径。通过Simulink与Carsim联合仿真,结果表明所改进的APF算法能规划出更加合理的路径,设计的LQR路径跟踪控制器跟踪误差趋近于零,能够较好地控制车辆完成超车换道。

Dynamic path planning for intelligent vehicle overtaking based on improved potential field method

In order to improve the problems of poor trajectory specification, unsecured safety, and unsmooth paths planned by the artificial potential field method (APF) in the high-speed overtaking state of intelligent vehicles. By designing the lane-changing gravitational potential field used to guide the smart vehicle to change lanes in order to ensure the normality of the lane-changing trajectory; constructing an obstacle vehicle repulsive potential field with an anisotropic distribution structure to improve the safety of lane changing; and establish the road centerline repulsive potential field with steering notch and design the smoothing cost function to optimize the planned trajectory points to make the planned path smoother. Design of path tracking controller based on linear quadratic regulator (LQR) to track the planned path. Through the joint simulation of Simulink and Carsim, the results show that the improved APF algorithm can plan a more reasonable path, and the tracking error of the designed LQR path tracking controller tends to be close to zero, which can better control the vehicle to complete the overtaking lane change.