分布式完全覆盖及探索的多机器人系统  

Distributed Multi-Robot Simultaneous Coverage and Exploration System

作　　者：刘永凯 董伟[1] 

机构地区：[1]上海交通大学机械与动力工程学院,上海200240

出　　处：《机电一体化》2023年第1期62-67,共6页Mechatronics

摘　　要：机器人覆盖路径规划主要分为完全覆盖和探索两种任务,而在同时进行完全覆盖及探索两种任务的场景中,需另外设计框架。本文针对此种完全覆盖探索任务,研究多机器人完全覆盖探索路径规划框架,通过两个系统提升整体任务鲁棒性。首先,本文提出分布式机器人系统用以解除对中心式控制器的依赖,其二,本文采用神经网路压缩点云地图,提高传输效率,使得地图能够在线传输合并,解决途中机器人故障地图丢失之问题。故本研究提出完全覆盖探索系统,既能解决依赖中心控制的问题,还能处理部分机器人故障的丢包,并另借由机器人仿真环境验证本文之框架的有效性。Robot coverage path planning can be divided into two main tasks:complete coverage and exploration,but in the scenarios where both complete coverage and exploration tasks are carried out,a separate framework needs to be designed.This paper is aiming to design a robust system in this kind of scenarios.First,we propose a distributed robot system to release the dependence on the central controller,and second,we adopt a neural network to compress and decompress the point cloud map to improve the transmission efficiency,so that the map can be forwarded and merged online to solve the problem of losing the map due to robot failure on the way.The validity of our framework of is also verified by using robot simulation environment.

关 键 词：机器人 区域覆盖 路径规划 分布式系统 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]