摘要
通过使用开源的机器人操作系统（ROS）,设计了采用SICK激光雷达的自动导引运输车（AGV）导航系统,系统包括硬件和软件两部分。硬件部分,采用ARM Cortex—M4内核的STM32F407处理器为电路核心,利用惯性测量单元（IMU）、轮式编码器以及激光雷达作为导航数据源,进行了相关的硬件设计。软件部分的路径规划模块,针对目前A＊算法应用于工厂等大规模环境时,未考虑AGV宽度信息和轨迹的问题,结合奖惩机制提出了改进的RA＊算法,有效地实现优先生成与原路径同方向的节点,避免生成对角线方向上与障碍物邻近的节点,通过ROS的plugin技术实现了该算法。最终在先锋机器Pioneer3-DX上,搭建本文设计的系统,检验了改进的RA＊算法在直线和L型走廊环境中导航效果。实验结果表明,改进的RA＊算法更符合工厂环境对AGV的需求。
In this paper,we use robot operation system and SICK laser radar to design an AGV navigational system.The system consists of two parts：hardware and software.The STM32 F407 processor with ARM Cortex-M4 kernel is the core-circuit.The inertial measurement unit（IMU）,the wheel encoder and the laser radar are used as navigation data source.In view of A＊ algorithm,in a large-scale environment,such as factories,it can＇t run efficiently,stably and steadily.Without considering the minimum turning radius and trajectory characteristics of the AGV motion in the factory environment and the width information of AGV,we proposed Improved RA＊ algorithm combined with reward and penalty mechanism.It generates node in the same direction as the original path,so as to avoid generating node adjacent to the obstacles in the diagonal direction.Finally,we build this AGV system on Pioneer3-DX and applied this algorithm to the corridor environment.The experimental results show that Improved RA＊ algorithm meets the requirements of the factory environment for AGV.
作者
韩昊旻
张崇明
陈志红
李传江
Han Haomin;Zhang Chongming;Chen Zhihong;Li Chuanjang(Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, Chin)
出处
《电子测量技术》
2018年第8期112-117,共6页
Electronic Measurement Technology
基金
上海市科委科技专项(16070502900)
上海师范大学创新团队项目(A-7001-15-001005)
上海师范大学校级科研项目(SK201718)项目资助
关键词
ROS
激光雷达
AGV
导航系统
RA＊算法
robot operation system
laser radar
automated guided vehicle
navigational system
RA ＊ algorithm
