Density-Based Optimal UAV Path Planning for Photovoltaic Farm Inspection in Complex Topography

Yifan Ding,Rui Cao,Siming Liang,Fengyang Qi,Qiang Yang,Wenjun Yan

DOI: https://doi.org/10.1109/ccdc49329.2020.9164257

2020-01-01

Abstract:In order to improve power generation efficiency and reduce land cost, many large-scale photovoltaic (PV) farms are built on the mountain. The rugged terrain of the mountains leads to an irregular distribution of the PV modules' location. Compared with manual inspection, the efficiency of unmanned aerial vehicles (UAVs) inspection is much higher. The path planning of scattered PV modules has become a challenge. In this paper, we use improved Density-Based Spatial Clustering of Applications with Noise (DBSACN) to partition the entire area. In order to reduce the number of turns of UAV, the partitions tend to divide the same row of PV modules into the same area. For the PV modules in each area, the full coverage path planning (CPP) of the boustrophedon will be adopted, and the lateral spacing between the PV modules can fly directly. Finally, the path control points are smoothed using Bezier Curves interpolation.