基于改进蚁群优化算法的自动驾驶多车协同运动规划

当前面向多辆自动驾驶汽车的协同运动规划方法能有效保证运行车辆与障碍物及其他车辆之间避免发生碰撞并保持安全距离，但车辆间的在线协同与规划能力仍有待提升。为实现多辆自动驾驶汽车在运动过程中的协同控制，提出一种基于改进蚁群优化算法的多车在线协同规划方法。以空间协同与轨迹代价为优化目标，构造多目标优化函数，确保了多车行驶过程中的协同安全性与轨迹平滑性。将多目标优化函数引入蚁群优化算法的信息素更新过程中，根据自动驾驶车辆数量产生多个种群，使得种群之间相互独立的同时为每辆自动驾驶汽车规划可行路线。最终对蚁群优化算法中的挥发因子进行自适应调整，提升了算法全局搜索能力及收敛速度。实验结果表明，该方法能使多辆自动驾驶汽车在运动过程中保持协同控制并规划出无碰撞路线，相比于基于人工势场和模型预测的协同驾驶方法在复杂道路场景下车辆间的协同效果更好且适应性更强。

Multi-Vehicle Collaborative Motion Planning for Autonomous Driving Based on Improved Ant Colony Optimization Algorithm

Current collaborative motion planning methods for multiple autonomous driving vehicles can effectively prevent collisions and maintain a safe distance between running vehicles and obstacles and other vehicles.However, the online collaboration and planning capabilities between vehicles still need to be improved.This study proposes a multi-vehicle online collaborative planning method based on an improved Ant Colony Optimization(ACO) algorithm to achieve the collaborative control of multiple autonomous driving vehicles during motion.By taking space collaboration and trajectory cost as optimization objectives, a multi-objective optimization function is constructed to ensure collaboration safety and trajectory smoothness during multi-vehicle driving.The multi-objective optimization function is introduced into the pheromone update process of the ACO algorithm, and multiple populations are generated based on the number of autonomous driving vehicles.The populations are independent and plan feasible routes for autonomous driving vehicles.The volatile factor in the ACO algorithm is adaptively adjusted, improving the global search ability and convergence speed of the algorithm.The experimental results show that the proposed method can maintain multiple autonomous driving vehicles in collaborative control during the movement process and plan a collision-free route.Compared with the collaborative driving method based on Artificial Potential Field(APF) and model prediction, the proposed method has a better collaborative effect between vehicles and more robust adaptability in complex road scenes.