UAV Coverage Path Planning of Multiple Disjointed Regions Based on Variable Neighborhood Search Algorithm

Coverage path planning (CPP) is crucial when unmanned aerial vehicle (UAV) performs tasks such as pesticide spraying. However, existing coverage path planning methods suffer from the problem of easily falling into local optimality. Therefore, an optimization method based on Variable Neighborhood Search (VNS) algorithm is proposed in this paper. First, an integer programming model is established for the problem, and then the problem is solved in two stages - 1) calculating the set of four optimal coverage paths for each mission area and applying cubic Bezier curve for smoothing the turning paths; 2) generating initialized coverage path and iteratively optimizing them using the VNS algorithm, which has a neighborhood structure includes three kinds of 2-OPT, 3-OPT and Segment Shift to enhance the local search capability of the algorithm and minimize the total cost of the coverage flight. Simulation results show that the proposed method exhibits high performance in instances of different scales, demonstrating the superiority of the method in solving the task of covering multiple disjointed regions.