Aiming at the problems of poor initial population quality,insufficient population diver-sity and unsatisfactory path length of traditional genetic algorithm,an improved genetic algo-rithm is proposed.By generating the initial path based on the gravitational field model,the initial population quality is improved;The penalty factor and incentive factor are added to the fitness function to enhance the population quality screening;The differential evolution algorithm is in-troduced to vectorize the differences between the individuals of the populations,and the number of mutations of the populations is controlled by the mutation probability to optimize the diversity of the populations,so as to get the global optimal solution in a better and faster way.Improved genetic algorithm,traditional genetic algorithm and ant colony algorithm are used to carry out simulation experiments on different raster maps for path planning,and the results show that the improved genetic algorithm of this paper can quickly find the optimal path when dealing with this kind of path planning problem,and the optimal path is shortened by 17.39%and 7.9%in the environment of M3 map with high complexity,compared with the traditional genetic algorithm and the ant colony algorithm,respectively.