無人飛行体によるインフラ設備外観自動点検システムのための状態データの精度と作業効率を考慮した計測位置決定

抄録

The authors have researched into path-planning of Unmanned Aerial Vehicles (UAVs) for automatic visual inspection system. It is a important function for inspections to ensure accuracy and consider efficiency of operations in data-gathering to detect anomalies. In this research, it is assumed that the 3D model of a target facility, required accuracy of gathering data and specs of instruments are given. Under this assumption, in our previous method, the measuring positions were calculated systematically, which means the positions of UAVs to gather the accurate data of the target facility were calculated. Then, by setting the measuring positions as waypoints of UAVs, a general path-planning method was used to plan the flight path which is appropriate for visual inspection. Although accuracy of operations was ensured, efficiency could not be considered sufficiently. This paper presents a new path-planning method ensuring accuracy and considering efficiency. In our new method, the triangular polygons for shape description of the 3D model are reconstructed using k-means++ clustering in order to ensure accuracy and consider efficiency in calculation of measuring positions. The proposed method is compared with the previous method through computer simulations using the 3D models of two typical bridges: a girder bridge and an arch bridge.