Research on shortest path planning for unmanned aerial vehicles based on genetic algorithm

A genetic algorithm based on roulette selection method and integer crossover operator is proposed for the UAV path planning problem. First, this paper introduces the application background and historical origin of the path planning problem. Subsequently, this paper elaborates the mathematical model of the UAV path planning problem and illustrates the UAV path planning problem with graphical examples. The algorithm introduces a roulette selection operator and an integer crossover operator to ensure that the algorithm explores more solution space during the search process, thus increasing the chance of finding a globally optimal solution. The variation operator adopts the strategy of local perturbation and retaining excellent individuals, which makes it possible for the algorithm to break through the local extremes towards a globally better evolution. In the simulation experiments, the path planning problems of 30 cities, 51 cities and 76 cities are solved respectively, and their trajectory graphs and iteration curves are obtained. The experimental results show that the algorithm has good performance in solving the UAV path planning problem.