基于混合算法的电力杆塔巡检实时航迹规划

Real-Time Path Planning for Power Tower Inspection Based on Hybrid Algorithm

传统的电力杆塔拍摄视点顺序固定,多旋翼无人机巡检距离并非最优;同时,随着维度增加,航迹规划算法空间复杂度呈指数增长,不能满足实时规划航迹的需求。针对以上问题,提出一种基于蚁群和A*混合算法(ACO-A*)的电力杆塔巡检三维航迹规划方法。该方法分为全局规划和局部规划,全局规划利用改进蚁群算法找到覆盖所有视点的较优路径,并通过算法判断路径是否经过障碍物,再运用A*算法局部规划。仿真结果表明:ACO-A*算法规划的航迹长度比《架空输电线路无人机巡检影像拍摄指导手册》规定的巡检航迹降低了16.85%;ACO-A*算法路径规划时间比A*算法降低了99.68%。因此本方法既节约了巡检能耗,又提高了航迹规划的效率。

The sequence of conventional shooting viewpoints for power tower is fixed and the inspection distance of multi-rotor UAV is not optimal.In addition,as the dimension increases,the path planning algorithm cannot meet the requirements of real-time path planning because the space complexity increases exponentially.Aiming at those problems,a three-dimensional path planning method for power tower inspection is proposed based on ant colony optimization and A*(ACO-A*)hybrid algorithm.The method is composed of global planning and local planning.Firstly,the global planning uses the ant colony optimization algorithm to find a relatively optimal path that covers all viewpoints,and to judge whether the path passes through obstacles.And then the A*algorithm is used for local planning.The simulation results show that the path length planned by the proposed ACO-A*algorithm is reduced by 16.68%compared to that stipulated in the Shooting Manual for UAV Inspection Images of Overhead Transmission Lines,and the path planning time is reduced by 99.68%compared to that of the A*algorithm.Therefore,the proposed method not only reduces the energy consumption for inspection,but also enhances the efficiency of path planning.