蚁群优化算法
运动规划
人工蜂群算法
计算机科学
路径(计算)
算法
启发式
趋同(经济学)
数学优化
路径长度
管道(软件)
移动机器人
人工智能
蚁群
机器人
数学
经济
经济增长
程序设计语言
计算机网络
作者
Guangxin Li,Chao Liu,Lei Wu,Wensheng Xiao
标识
DOI:10.1016/j.asoc.2023.110868
摘要
Path planning is involved in many applications such as trajectory planning, mobile robotics, pipeline layout, etc. Researchers use artificial intelligence algorithms to solve path planning efficiently, among which the ant colony algorithm (ACO) is one of the common intelligent algorithms to solve path planning problems. However, the traditional ACO has defects such as low early search efficiency and easy to fall into local optimum, while the artificial bee colony algorithm (ABC) has high search efficiency. Therefore, an improved ant colony optimization-artificial bee colony algorithm (IACO-IABC) is proposed in this study. IACO-IABC contains three mechanisms. First, the heuristic mechanism with directional information for the ACO is improved to enhance the efficiency of steering towards the target direction. Secondly, the novel neighborhood search mechanism of the employed bee and the onlooker bee in the ABC is presented to enhance the exploitation of optimal solutions. Then, the path optimization mechanism is introduced further to reduce the number of turn times in the planned path. To verify the performance of the IACO-IABC, a series of experiments are conducted with 10 different maps. The experiments compare nine variants of ACO and eight commonly used intelligent search algorithms, and the results show the advantages of the IACO-IABC in reducing the number of turn times and path lengths and enhancing the convergence speed of the algorithm. Compared to the best results of other algorithms, the average improvement percentages of the proposed algorithm in terms of the path turn times are 375%, 258.33%, 483.33%, 186.67%, 166.77% and 255.33%, further demonstrating the ability of IACO-IABC to obtain high-quality path planning result.
科研通智能强力驱动
Strongly Powered by AbleSci AI