基于A*和改进动态窗口法的建筑移动机器人路径规划

Path Planning of Construction Mobile Robot Based on A*and Improved Dynamic Window Approach

针对基于麦克纳姆轮移动底盘的建筑移动机器人在建筑施工作业场景下存在的路径偏离全局安全路径过大等问题,提出一种基于A*和改进动态窗口法的路径规划算法。首先,根据麦克纳姆轮运动全向性的特点,优选A*算法中节点扩展的方向和启发函数,得到全局最优路径;其次,在构建速度窗口和建立轨迹预测模型时,只考虑线速度,并结合全局路径和滑移角以及航向角信息对评价函数进行优化,提高运动速度的平滑性,减少跟踪误差;最后,实验结果表明,改进后降低了全局路径跟踪误差,保证了运动的稳定性和安全性。验证了该文所提算法的有效性。

Aiming at the problems of the movement path deviates too much from the global safe path of the construction mobile robot based on the mecanum wheel mobile platform,a path planning algorithm based on Aand improved dynamic window approach is proposed in this paper.Firstly,according to the characteristics of omnidirectional mobile platform based on the mecanum wheel,the direction and heuristic function of node expansion in the Aalgorithm are optimized,and the global optimal path is obtained.Second,the velocity search space and forward predictive kinematic equations containing only linear velocity are established,and the evaluation function is optimized by combining global path,slip angele and heading informations.Finally,the experimental results show that after the improvement,the global path error is reduced,the stability and safety of the movement are ensured.The effectiveness of the algorithm proposed in this paper is verified。