Area Coverage Path Planning for Fixed-Wing Unmanned Aerial Vehicle Based on Genetic Algorithm Incorporating Simulated Annealing Mechanism

An improved genetic algorithm incorporating simulated annealing operations is proposed to address the coverage path planning problem for fixed-wing unmanned aerial vehicles (UAVs). After generating the back-and-forth paths (BFPs) in the width direction of the convex polygonal task area, the sequence of BFPS is optimized by transforming the problem into a traveling salesman problem (TSP). The initial population is constructed using a greedy algorithm. The mutation intensity varies with the iteration number, and during the selection process, individuals that are both high-quality and have a genetic distance greater than the elitist retention threshold are retained. Following crossover, mutation, and selection operations, an enhanced simulated annealing process is applied to a subset of the offspring individuals. The performance of the proposed algorithm is validated through simulation experiments within the convex polygonal area. Results demonstrate that the algorithm exhibits significant performance advantages in terms of shorter path length and lower computational cost compared to other algorithms, especially when the number of BFPs is large.