An Adaptive Improved Ant Colony System Based on Population Information Entropy for Path Planning of Mobile Robot

In this paper, an adaptive improved ant colony algorithm based on population information entropy(AIACSE) is proposed to improve the optimization ability of the algorithm. The diversity of the population in the iterative process is described by the information entropy. The non-uniform distribution initial pheromone is constructed to reduce the blindness of the search at the starting phase. The pheromone diffusion model is used to enhance the exploration and collaboration capacity between ants. The adaptive parameter adjusting strategy and the novel pheromone updating mechanism based on the evolutionary characteristics of the population are designed to achieve a better balance between exploration of the search space and exploitation of the knowledge during the optimization progress. The performance of AIACSE is evaluated on the path planning of mobile robots. Friedman's test is further conducted to check the significant difference in performance between AIACSE and the other selected algorithms. The experimental results and statistical tests demonstrate that the presented approach significantly improves the performance of the ant colony system (ACS) and outperforms the other algorithms used in the experiments.

Automated summary

The AIACSE algorithm aims to improve optimization ability by describing population diversity through information entropy and reducing search blindness through non-uniform initial pheromones. It uses a pheromone diffusion model to enhance exploration and collaboration among ants, with adaptive parameter adjustment and a novel pheromone updating mechanism for better balance between exploration and exploitation. Experimental results show significant performance improvement in path planning and demonstrate superiority over other algorithms.