Aiming at the problems of slow algorithm convergence and insufficient ability to approach the optimal solution in the ant colony algorithm, this paper proposed an ant colony algorithm based on heterogeneous dual-population and global vision, and applied to the field of mobile robot path planning. Firstly, the algorithm studied the parallel structure based on heterogeneous ant colonies, improved the convergence speed and the ability to plan the optimal path by collaboration through differentiated populations. Then, the algorithm studied the adaptive step size with global field of view, and solved the problem that the ant colony algorithm could not search for the optimal step size due to the local field of view. Finally, the algorithm studied the pheromone initialization and updated methods, improved the search disorder and unreasonable pheromone update in the initial stage of the traditional ant colony algorithm. The experimental results show that the algorithm has a significant improvement over the comparison methods in terms of approaching the optimal solution ability and improving the convergence speed. In the several simulation maps tested, the average path length is optimized by 12%, and the average number of iterations and average running time are reduced by 67% and 82% respectively.

[2022-01-25] Accepted Paper /doc/2021.12.0632/v1

[2022-06-05] Printed Article /doc/2021.12.0632/v2