A self-adaptive genetic algorithm for the shortest path planning of vehicles and its comparison with Dijkstra and A^* algorithms

A self-adaptive genetic algorithm was proposed and successfully applied for the shortest path planning of vehicles. The encoding scheme, crossover and mutation operators were specifically designed for shortest path planning problems in the proposed genetic algorithm. A new online self-adaptive adjustment strategy for crossover and mutation probabilities was also investigated in order to improve the search speed and search quality of genetic algorithm. The comparison of the proposed genetic algorithm with Dijkstra and A^* algorithms was carried out. Simulation results under 5 different circumstances show that the proposed genetic algorithm can decrease the searching time for shortest path compared with Dijkstra algorithm and obtain more shortest paths than A^* algorithm.