计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (22): 305-312.DOI: 10.3778/j.issn.1002-8331.2102-0339

• 工程与应用 • 上一篇    

基于APF的AGV局部路径规划改进算法研究

丁承君,阎欣怡,冯玉伯,贾丽臻   

  1. 1.河北工业大学 机械工程学院,天津 300130
    2.天津通信广播集团有限公司 智慧云网事业部,天津 300143
    3.中国民航大学 航空工程学院,天津 300300
  • 出版日期:2022-11-15 发布日期:2022-11-15

Improved Algorithm of AGV Local Path Planning Based on APF

DING Chengjun, YAN Xinyi, FENG Yubo, JIA Lizhen   

  1. 1.School of Mechanical Engineering, Hebei University of Technology, Tianjin 300130, China
    2.Intelligence Industry Division, Tianjin Communication & Broadcasting Group Co., Ltd., Tianjin 300143, China
    3.School of Aeronautical Engineering, Civil Aviation University of China, Tianjin 300300, China
  • Online:2022-11-15 Published:2022-11-15

摘要: 针对原有人工势场法(artificial potential field,APF)在局部路径规划时的避障效果不良问题,提出一种APF-PSO的改进算法改善原算法优化路径规划的效果。将速度势场引入位置势场中使AGV(automated guided vehicle)动态避开不同速度的移动障碍物;当算法陷入局部最小值时,采取PSO(particle swarm optimization)算法,并对其惯性权重因子和学习因子做出调整,通过三次样条曲线插值来平滑路径,使得AGV找到最短路径。结果表明APF-PSO改进算法可根据障碍物速度不同动态避障,解决了APF算法运算中避障效果不良问题。

关键词: 局部路径规划, 人工势场法改进, 斥力, 局部最小值, 收敛效率

Abstract: Aiming at the problem of poor obstacle avoidance effect of the original APF(artificial potential field) algorithm in local path planning, an improved APF-PSO algorithm is proposed to improve the effect of original algorithm for optimizing path planning. The velocity potential field is introduced into the position potential field to make the AGV(automated guided vehicle) dynamically avoid moving obstacles of different speeds. When the algorithm falls into a local minimum, the PSO algorithm is adopted and its inertia weight factor and learning factor are adjusted, the path is smoothed by cubic spline interpolation, so that the AGV can find the shortest path. The results show that the improved APF-PSO algorithm can dynamically avoid obstacles according to different obstacle speeds, and solve the problem of poor obstacle avoidance effects in the calculation of APF algorithm.

Key words: local path planning, improvement of artificial potential field method, repulsion, local minimum, convergence efficiency