基于ROS的工业机器人及AGV的视觉协同控制系统

Visual Cooperative Control System for Industrial Robots and AGVs Based on ROS

当前自动化上下料的机器人之间仍各自独立运行,上下料位置固定缺乏柔性,智能化程度低。针对以上问题设计了基于ROS的工业机器人及AGV的视觉协同控制系统,用以实现智能化、柔性化的上下料工作。在ROS系统上,基于全局视觉与改进ArUco tag识别技术,开发了机械臂和AGV的识别与控制模块,实现了两种机器人在全局视觉下的识别与路径规划;并基于动态指令技术和ROS主从机通讯技术,实现了所开发ROS系统与机械臂和AGV的可靠通讯;最后,考虑不同机器人的协作规则和目标位姿计算方法,结合上述机器人识别与控制模块,提出了一种机械臂与AGV的视觉协同控制方法。实验测试表明,该系统能够实现多机器人的识别、控制与协同,工件托盘能够通过AGV按照要求实现任意指定位置的上下料接驳,具有良好的稳定性与柔性。

The current automatic loading and unloading robots still operate independently,the loading and unloading position is fixed without flexibility,and the degree of intelligence is low.Aiming at the above problems,a ROS-based visual cooperative control system of industrial robot and AGV was designed to realize intelligent and flexible loading and unloading work.On ROS system,global recognition and positioning of manipulator and AGV were realized based on global vision and improved ArUco tag recognition technology.The dynamic instructions generated by the system were sent to the manipulator through TCP/IP communication to control the movement of the manipulator.The control signal after path planning was sent to AGV through ROS master-slave communication to make AGV run along the target trajectory;The task was judged by the cooperative node,the target pose was calculated,and the target pose was sent to the manipulator and AGV control node according to the task requirements.The experimental test showed that the system can realize the identification,control and coordination of multiple robots,and the workpiece tray can realize the loading and unloading connection of any specified position according to the requirements of AGV,with good stability and flexibility.