Mobile robot path planning based on an improved ACO algorithm and path optimization
Tianfeng Zhou,Wenhong Wei
DOI: https://doi.org/10.1007/s11042-024-19370-x
IF: 2.577
2024-05-19
Multimedia Tools and Applications
Abstract:This paper will improve the fundamental ant colony optimization algorithm in the context of mobile robot path planning in response to its flaws, which include easy descent into a local optimum, a large number of inflection points along the path, and an inefficient convergence speed. Based on the two-dimensional grid map modeling approach, map pheromone initialization is unevenly assigned to improve the blindness of the ant colony optimization algorithm in the initial phase of path planning. Furthermore, an adaptive pheromone volatility factor is proposed to improve the global characteristics of early path planning, accelerate the convergence speed as the iteration advances, and avoid a local optimum that could impede the algorithm's progress. The pheromone update mechanism was redesigned, and a reward and punishment mechanism was incorporated into the pheromone update function to accelerate the convergence of the algorithm. Finally, path optimization diminishes the quantity of inflection points along the path, which makes the path smoother and shortens the path length. The outcome of the experiments demonstrate that the improved ant colony optimization algorithm performs better, converges more quickly, and more closely matches the needs of actual mobile robot walking.
computer science, information systems, theory & methods,engineering, electrical & electronic, software engineering