移动机器人全覆盖路径的BINN-元胞自动机规划  

Mobile Robot Complete Converge Path Planning Based on BINN-Cellular Automata System

作　　者：朱方园[1] ZHU Fang-yuan(School of Intelligent Manufacturing,Nanjing Vocational College of Information Technology,Jiangsu Nanjing 210023,China)

机构地区：[1]南京信息职业技术学院智能制造学院,江苏南京210023

出　　处：《机械设计与制造》2024年第8期346-349,共4页Machinery Design & Manufacture

基　　金：2016年度江苏省高校自然科学研究面上资助经费项目(16KJB470019)。

摘　　要：为了实现机器人对工作区域的全覆盖,提出了基于生物激励神经网络-元胞自动机系统的全覆盖路径规划方法。介绍了生物激励神经网络算法的基本原理,分析了该算法在机器人陷入死区时无法逃逸的问题。基于元胞自动机系统设计了机器人逃逸机制,包括逃逸点的确定和逃逸路径的规划方法。在仿真环境下,将元胞系统逃逸机制与基本RRT、文献[10]的BINN-RRT逃逸机制进行对比,结果表明元胞系统逃逸机制的规划时间比基本RRT小2个数量级,比BINN-RRT小1个数量级,且逃逸路径短于另外两种方法,验证了元胞系统逃逸机制的有效性和优越性。基于BINN和元胞系统的全覆盖路径比BINN-RRT规划路径更加平滑,验证了全覆盖方法的优越性和有效性。In order to realize the full coverage of the robot working area,a path planning method based on biological excitation neural network cellular automata system is proposed.This paper introduces the basic principle of the biological excitation neural network algorithm,and analyzes the problem that the algorithm cannot escape when the robot falls into the dead zone.Based on cellular automata system,the escape mechanism of robot is designed,including the determination of escape point and the planning method of escape path.In the simulation environment,the escape mechanism of cellular system is compared with basic RRT and BINN-RRT escaping mechanism in reference[10].The results show that the planning time of escape mechanism of cellular system is two orders of magnitude less than basic RRT and one order of magnitude less than BINN-RRT,and the escape path is shorter than the other two methods,which verifies the effectiveness and superiority of escape mechanism of cellular system.The full coverage path based on BINN and cellular system is smoother than that based on BINN-RRT,which verifies the superiority and effectiveness of the full coverage method.

关 键 词：移动机器人 全覆盖路径规划 生物激励神经网络 元胞自动机系统 逃逸机制 

分 类 号：TH16[机械工程—机械制造及自动化] TP242[自动化与计算机技术—检测技术与自动化装置]