基于学习分类器的自主地面车在狭隘环境中的路径规划

Autonomous Land Vehicle Path Planning Based on Learning Classifier System in Narrow Environments

提出了一种基于学习分类器(LCS)的避碰路径规划方法,设计了集成适应度函数,在确保安全避碰的前提下,解决自主地面车(ALV)在狭隘环境下的路径优化问题.不同环境的仿真实验结果表明,遗传算法和学习分类器结合用于自主地面车的路径规划是收敛的,提高了ALV在狭隘环境中快速发现安全路径的能力.

A collision avoidance path planning method based on LCS（learning classifier system） is present,and an integrated fitness function to solve ALV＇s（autonomous land vehicle） path optimization problem is designed in the narrow environment under safe collision avoidance.Different environment simulation results show that ALV＇s path planning is convergent by combining genetic algorithms and learning classifier system,and ALV＇s capabilities of quickly finding the secure path in the narrow environments is improved.