摘要
为了实现移动机器人在未知环境下快速自主定位与建图(SLAM),以及精准的路径规划与避障等功能。文中以Jetson nano作为主控,SMT32作为下位机,以深度相机RGB-D、高精度惯性测量传感器IMU、激光雷达等为传感器搭建自主移动机器人的硬件系统;控制系统以ROS系统为软件基础,提出了一种“RGB-D+IMU+激光雷达”多传感融合SLAM策略,运用ORB-SLAM3算法实现快速实时定位与建图(SLAM),采用Dijkstra算法为全局路线规划算法、DWA算法为局部路线规划算法实现路径规划和避障。最后在现实环境进行实验,从实验结果来看,达到了预期效果。
In order to realize the functions of rapid Autonomous Simultaneous Localization and Mapping(SLAM),accurate path planning and obstacle avoidance of mobile robots in unknown environments,this paper uses Jetson nano as the main control,SMT32 as the lower computer,and uses the depth camera RGB-D,inertial navigation sensor IMU,laser radar and other sensors to build the hardware system of the autonomous mobile robots.The control system is based on ROS system,and a"RGB-D+IMU+lidar"multi-sensor fusion SLAM strategy is proposed,and the ORB-SLAM3 algorithm is used to realize fast real-time Simultaneous Localization and Mapping(SLAM).Dijkstra algorithm is used as the global route planning algorithm,and DWA algorithm is used as the local route planning algorithm to realize path planning and obstacle avoidance.Finally,the experiment is carried out in the real environment,and the results show that the expected effect is achieved.
作者
黄广伟
陈梦婷
廖伟涛
HUANG Guangwei;CHEN Mengting;LIAO Weitao(College of Intelligent Manufacturing,Dongguan City University,Dongguan 523419,China)
出处
《机械工程师》
2023年第12期15-17,22,共4页
Mechanical Engineer
基金
东莞城市学院青年教师发展基金“基于视觉Slam的自主移动机器人多传感融合研究”(2021QJY005Z)
东莞市社会发展科技项目“基于深度学习与深度相机的机器人智能分拣关键技术研究”(20211800900572)
东莞城市学院高等教育教学改革项目“新工科背景下‘工业机器人技术基础’实验教学改革与探索”(2021yjjg015)。
