Flight path planning based on an improved genetic algorithm

Flight path planning for UAV is a complicated optimization problem with multiple constrains. In this paper, an improved dual-population genetic algorithm (IDPGA) is proposed. It uses an additional population to maintain population diversity of genetic algorithm (GA). The two populations have different evolutionary objectives and thus use different fitness functions. Generating offspring of each population is performed by randomly generating new individuals, inbreeding between individuals in the same population and crossbreeding between individuals from different populations. The next generation is produced by selecting the best ones from current populations and offspring. Besides, in order to improve the convergence performance of the algorithm, the initial populations are generated based on multiple constraints. The experimental results show that IDPGA improves the global search and local search capabilities for GA to ensure the global optima of the flight path.