面向动态避障的智能汽车滚动时域路径规划

为解决城市低速条件下智能汽车在避障过程中的路径规划问题,提出面向动态避障的智能汽车滚动时域路径规划方法。首先,划分车道可行区域,利用3次拉格朗日插值法拟合车道边界,并根据"车-路"的相对位置关系将车道区域进一步划分为车道间区域与车道内区域两部分。其次,以区域虚拟力场进行动态交通场景模拟,包括在障碍车周身沿车道方向的虚拟矩形区域斥力场,行驶目标位置的虚拟引力场和车道保持虚拟区域引力场3个部分,然后结合划分的车道区域确定各虚拟力场的作用区域。再次,建立主车动力学与运动学模型,障碍车运动学预测模型,把主车与障碍车无碰撞,主车行驶在车道内区域,趋向目标位置以及保证车辆稳定性作为优化目标,综合车辆模型的控制输入、状态变量等动力学约束条件,构建多目标的滚动时域控制器用于车辆避障路径规划,求解获得前轮转角作为控制量。最后,利用MATLAB和veDYNA软件对提出的路径规划控制系统分别在静态障碍和动态障碍工况下进行联合仿真。研究结果表明：该方法能够很好地解决躲避静态障碍和低速动态障碍车的问题,控制车辆驶向目标位置,并且在避障过程中满足车辆的动力学约束,同时又不会与道路边界发生碰撞,保证了车辆的安全性和稳定性。

Moving Horizon Path Planning for Intelligent Vehicle Considering Dynamic Obstacle Avoidance

To solve the slow speed path planning problem of intelligent vehicles in the process of obstacle avoidance in the urban environment, a moving horizon path planning method for dynamic obstacle avoidance is proposed. First, the Lagrange interpolation method was used to fit the lane boundary, and the lane was further divided into two parts:the inter lane area and the lane area according to the relative position of the "car-road"(Location of the vehicles on the road and relative position between vehicles). Second, the virtual force field of the region was established, including the virtual rectangular repulsion area along the lane of the obstacle vehicle, the virtual gravitational field of the target position, and the gravitational field region for lane keeping, and then the functional area of the virtual force field was confirmed according to the divided area. Third, the dynamics and kinematics model of the host vehicle and kinematics model of the obstacle vehicle were established. No collision between the host vehicle and the obstacle vehicle, the host vehicle driving in the lane area, orientation of the target position, and vehicle stability were taken as optimization objections. Then, considering the dynamic constraints of the input variables and the state variables, a multi-objective moving horizon controller was constructed to find the optimal front wheel angle for obstacle avoidance path planning. Finally, the proposed path planning control system was simulated under static obstacle and dynamic obstacle conditions via MATLAB and veDYNA. The simulation results show that the method can be used to avoid a static obstacle and low-speed dynamic obstacle vehicle by guiding the vehicle to the target position. Additionally, the dynamic constraints of the vehicle are satisfied during the journey without collision with the road boundary, which ensures the safety and stability of the host vehicle.