基于林地综合代价地图与开拓优化机制的改进蚁群算法
Improved Ant Colony Algorithm Based on Forest Comprehensive Cost Map and Research-optimization Mechanism
何浩天 1彭富明 1方斌 1相福磊 1张少杰1
作者信息
- 1. 南京理工大学 自动化学院,江苏 南京 210094
- 折叠
摘要
针对传统蚁群算法在林业机器人路径规划中未能考虑林区地形与地表状态,且收敛速度慢,易陷入局部最优的问题,提出一种基于林地综合代价地图与开拓优化机制的改进蚁群算法.基于林区地形地表构建林地综合代价地图;在蚁群搜索阶段中引入开拓优化机制;在轮盘赌中引入随机游走机制,并将其应用到林业机器人路径规划问题中.实验表明:改进蚁群算法能搜寻到更佳路径,具有更好的全局寻优能力.
Abstract
An improved ant colony algorithm based on the comprehensive cost map of forest land and the research-optimization mechanism is proposed to address the problem of the traditional ant colony algorithms which is absent of considerationof forest terrain and surface conditions in forestry robot path planning,slow at convergence speed and prone to falling into local optima.On the terrain and surface of the forest area,a comprehensive forest land cost map is constructed;a pioneering optimization mechanism is introduced in the ant colony search stage;and a random walk mechanism is led into the roulette wheel mechanism,which is applied the path planning problem of forestry robots.The experiments show that the proposed algorithm proposed can search for better paths and has better exploration and optimization capabilities.
关键词
蚁群算法/精英奖励/双层蚁群算法/林间路径规划Key words
ant colony/elite rewards/double layer ant colony algorithm/forest path planning引用本文复制引用
基金项目
国家重点研发计划项目(2021YFE0194600)
江苏省科技计划项目(BZ2023023)
出版年
2024
国家科技期刊平台
NSTL
万方数据