基于改进双层蚁群算法的移动机器人路径规划?
作者:
作者单位:

新疆大学

作者简介:

通讯作者:

中图分类号:

TP242.6

基金项目:

国家自然科学基金项目(面上项目,重点项目,重大项目):(62063033,U1813220)


Mobile Robot Path Planning Using Improved Double-layer Ant Colony Algorithm?
Author:
Affiliation:

XinJiang university

Fund Project:

The National Natural Science Foundation of China (General Program, Key Program, Major Research Plan):(62063033,U1813220)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为提升移动机器人的路径规划能力,提出了一种改进双层蚁群算法,将蚁群划分为引导层蚁群和普通层蚁群. 为提升算法的收敛速度和路径的平滑程度,在设计引导层蚁群启发函数时加大终点栅格的吸引力, 设计普通层蚁群启发函数时同时考虑起点、终点和转折点的影响;针对复杂环境下蚁群算法死锁严重的问题,为引导层蚁群设计了应对死锁问题的自由寻路-剪枝策略,当引导层蚁群发生死锁时可以迅速跳出并优化路径;为进一步提升算法的运行效率,每一次迭代后仅对路径长度较短的路径进行信息素更新,并在信息素更新公式中引入次优路径抑制因子,充分发挥不同层蚁群在搜索过程中的协作优势,避免在迭代过程中陷入局部最优. 通过仿真实验验证了所提算法在大规模环境及复杂障碍环境的可行性、高效性和鲁棒性.

    Abstract:

    An improved double-layer ant colony optimization algorithm is proposed for mobile robot path planning.This double-layer ant colony optimization algorithm consists of a guiding layer and a common layer. First,the heuristic function for guiding layer increases the attractiveness of the ending point to accelerate the convergence speed,and then,the influence of starting point, ending point and turning point is considered to design heuristic functions of common layer for high search efficiency and smoothness. Besides,a freedom pathfinding-pruning method is designed for the guiding layer to solve the problem of deadlock in the complex environment, so that the guiding layer ant colony can avoid deadlock and optimize paths. Finally, a inhibited factor is applied to pheromone update rule and only the ant to find the high ranked path to the current loop is allowed to update the pheromone, which can further exploit the collaborative advantages of the double-layer ant colony in the search process and avoid running into the local optimum.Simulation results show that the proposed algorithm is more effective and robust in complicated and large environment.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-05-22
  • 最后修改日期:2020-11-30
  • 录用日期:2020-12-03
  • 在线发布日期: 2021-01-04
  • 出版日期: