基于单舵轮搬运机器人的导航系统设计  

Navigation system design based on single wheel handling robot

作　　者：于镭[1] 张国强 王泽龙 Yu Lei;Zhang Guoqiang;Wang Zelong(College of Automation and Electronic Engineering.Qingdao University of Science and Technology,Qingdao 266042,China)

机构地区：[1]青岛科技大学自动化与电子工程学院,青岛266042

出　　处：《电子测量技术》2020年第18期11-16,共6页Electronic Measurement Technology

摘　　要：针对单舵轮搬运机器人在仓库环境中如何定位和导航这一问题,设计了一款搬运机器人的室内导航系统。通过Rplidar-A2激光雷达传感器和Rock Pi-4A核心控制器以及自主研发的智能控制器完成了系统的硬件设计。通过获取激光雷达和IMU惯性测量单元等传感器数据编写了导航程序,在ROS分布式框架下完成了系统的软件设计。研究了基于扫描匹配的即时定位与建图(SLAM)算法,并在此基础上提出一种自适应粒子数增减的Gmapping算法。实验结果表明,优化后的Gmapping算法不但降低了建图时间消耗而且建立的地图精度较高,设计的导航系统有着良好的导航精度,能够满足实际工厂环境下的导航需求。To address the problem of how to locate and navigate a single steering wheel handling robot in a warehouse environment, an indoor navigation system for the handling robot is designed. The hardware design of the system was completed by the Rplidar-A2 LIDAR sensor and the Rock Pi-4 A core controller as well as an in-house developed intelligent controller. A navigation program was written by acquiring sensor data such as LIDAR and IMU inertial measurement unit, and the software design of the system was completed in the ROS distributed framework. A scan-matching based simultaneous localization and mapping(SLAM) algorithm is investigated and an adaptive Gmapping algorithm for increasing or decreasing particle number is proposed on this basis. The experimental results show that the optimized Gmapping algorithm not only reduces the time consumption of map building but also creates a more accurate map, the designed navigation system has good navigation accuracy and can meet the real factory environment navigation needs.

关 键 词：系统设计 ROS SLAM 激光雷达 Gmapping优化算法 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]