Multi-strategy Adaptable Ant Colony Optimization Algorithm and Its Application in Robot Path Planning
Junguo Cui,Lei Wu,Xiaodong Huang,Dengpan Xu,Chao Liu,Wensheng Xiao
DOI: https://doi.org/10.1016/j.knosys.2024.111459
IF: 8.139
2024-01-31
Knowledge-Based Systems
Abstract:As a widely used path planning algorithm, the ant colony optimization algorithm (ACO) has evolved into a well-developed method within the realm of optimization algorithms and has been extensively applied across various fields. In this study, a multi-strategy adaptable ant colony optimization (MAACO) is proposed to alleviate the insufficient and inefficient convergence of ACO, employing four-design improvements. First, a direction-guidance mechanism is proposed to improve the performance of node selection. Second, an adaptive heuristic function is introduced to decrease the length and number of turns of the optimal path solutions. Moreover, the deterministic state transition probability rule was employed to promote the convergence speed of ACO. Finally, nonuniform pheromone initialization was utilized to enhance the ability of ACO to select advantageous regions. Subsequently, the major parameters of the strategies were optimized and their effectiveness was validated. MAACO was proposed by combining these four strategies with ACO. To verify the advantages of MAACO, five representative environment models were employed, and comprehensive experiments were conducted by comparing them with existing approaches, including the A* algorithm, variants of ACO, Dijkstra's algorithm, jump point search algorithm, best-first search, breadth-first search, trace algorithm, and other excellent algorithms. The experimental statistical results demonstrate that MAACO can efficiently generate smoother optimal path-planning solutions with lower length and turn times and improve the convergence efficiency and stability of ACO compared to other algorithms. The generated results of MAACO verified its superiority in solving the path-planning problem of mobile robots.
computer science, artificial intelligence