Research Area:  Metaheuristic Computing

Abstract:

In view of the shortcomings of traditional ant colony algorithm (ACO) in path planning of indoor mobile robot, such as a long time path planning, non-optimal path for the slow convergence speed, and local optimal solution characteristic of ACO, an improvement adaptive ant colony algorithm (IAACO) is proposed in this paper. In IAACO, firstly, in order to accelerate the real-time and safety of robot path planning, angle guidance factor and obstacle exclusion factor are introduced into the transfer probability of ACO; secondly, heuristic information adaptive adjustment factor and adaptive pheromone volatilization factor are introduced into the pheromone update rule of ACO, to balance the convergence and global search ability of ACO; Finally, the multi-objective performance indexes are introduced to transform the path planning problem into a multi-objective optimization problem, so as to realize the comprehensive global optimization of robot path planning. The experimental results of main parameters selection, path planning performance in different environments, diversity of the optimal solution show that IAACO can make the robot attain global optimization path, and high real-time and stability performances of path planning.

Keywords:  
Path planning optimization
ant colony algorithm
Meta-Heuristic

Author(s) Name:  Changwei Miao, Guangzhu Chen, Chengliang Yan, Yuanyuan Wu

Publisher name:  Elsevier

DOI:  10.1016/j.cie.2021.107230

Volume Information:  Volume 156, June 2021, 107230

Paper Link:   https://www.sciencedirect.com/science/article/abs/pii/S0360835221001340