Open AccessUAV Path Planning Based on an Improved Ant Colony Algorithm
Author(s) -
Hui Song,
Minghan Jia,
Yihang Lian,
Yu Fan,
Keshan Liang
Publication year - 2022
Publication title -
journal of electronic research and application
Language(s) - English
Resource type - Journals
eISSN - 2208-3510
pISSN - 2208-3502
DOI - 10.26689/jera.v6i2.3808
Subject(s) - path (computing) , ant colony optimization algorithms , heuristic , motion planning , mathematical optimization , path length , local optimum , algorithm , computer science , ant colony , any angle path planning , set (abstract data type) , process (computing) , mathematics , artificial intelligence , robot , computer network , programming language , operating system
Reviews and experimental verification have found that existing solution methods can be used to solve UAV path planning problems, but each approximate solution has its own advantages and disadvantages. For example, ant colony algorithm easily falls into the local optimum in the process of realizing path planning. In order to prevent too low pheromones on the longer path and too high pheromones in the shorter path, the upper and lower limits of pheromones as well as their volatile factors are set to avoid falling into the local optimum. Secondly, multi-heuristic factors are introduced, and the overall length of the path serves as an adaptive heuristic function factor that determines the probability of state transition, which affects the probability of ants choosing the corresponding path. The experimental results show that the path length planned by the improved algorithm is 93.6% of the original algorithm, and the optimal path length variance is only 14.22% of the original algorithm. The improved ant colony algorithm shortens the optimal path length and solves the UAV path planning problem in terms of local optima. At the same time, multiple enlightening factors are introduced to increase the suitability of UAV for complex environments and improve the performance of UAV.