基于连续可微采样的无人机多向视点规划  

Multi-directional viewpoints planning of UAV inspection based on continuous differentiable sampling

作　　者：吴华[1] 徐肖顺 白晓静[1] WU Hua;XU Xiaoshun;BAI Xiaojing(AMaze,School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学控制与计算机工程学院电力智能机器人工程技术研究中心,北京102206

出　　处：《计算机集成制造系统》2024年第3期1161-1170,共10页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金青年项目(62202164,62301220)。

摘　　要：无人机在各种工业基础设施的自主巡检中,需要为大型复杂待巡检对象设计大量的数据采集航点,以获取高质量的全覆盖影像数据。逐点飞行作业能耗高采集效率低,限制了无人机大规模的工业应用。为了优化飞行作业模式,有效提升巡检效率,针对已知待巡检复杂设备结构提出一种全新的多向视点规划算法——连续可微采样的无人机多向视点规划算法(MD-VPP)。该方法在满足全面覆盖巡视目标以及数据采集质量要求的前提下,大幅减少航点数量。首先,将云台俯仰角、偏航角和相机成像面积作为约束条件,构建视点采集连续可微的视图质量目标函数,优化求解候选视点集;然后利用贪心算法求解全覆盖待巡检对象的航点位姿。通过实验对比分析了针对不同巡检对象的航点数量和航点数减少率,相比领域内其他优秀方法,所提方法在确保全覆盖的前提下航点数量至少降低77%,极大地提升了无人机单次作业的数据采集效率。When Unmanned Aerial Vehicle(UAV)is applied to the inspection of electric equipment,it is very important to improve the inspection efficiency and the quality of inspection data.Thus,a new Multi-Directional View Point Planning algorithm(MD-VPP)for the inspection of known complex equipment structures was present,which could greatly reduce the number of waypoints while satisfying the requirement of comprehensive coverage inspection target and data acquisition quality.A view quality function of continuous and differentiable view acquisition was constructed with the pitch,yaw and camera projection of PTZ as constraints,and the candidate view sets was optimized,and then the waypoint pose of the coveraged object was solved by greedy algorithm.The number and reduction rate of waypoints under different patrol objects were compared and analyzed experimentally.Compared to other excellent methods in the field,the proposed method reduced the number of waypoints by at least 77%while ensuring full coverage and greatly improved the data acquisition efficiency of a single UAV operation.

关 键 词：多视角 视点规划 全覆盖 无人机巡检 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]