ROS环境下机械臂物体抓取技术研究  

Research on Manipulator Object Grasping Technology in ROS

作　　者：王晶航 韩江桂[1] 张文群[1] 孙颙琰 WANG Jinghang;HAN Jianggui;ZHANG Wenqun;SUN Yongyan(College of Power Engineering,Naval University of Engineering,Wuhan 430033)

机构地区：[1]海军工程大学动力工程学院,武汉430033

出　　处：《舰船电子工程》2022年第4期49-53,共5页Ship Electronic Engineering

基　　金：国家部委科研项目(编号:HJ20182C070236);军内科研项目(编号:417214422)资助。

摘　　要：针对机械臂在非结构环境中物体识别与抓取的不确定性问题,对机械臂运动规划方法和物体识别抓取在ROS环境下开展仿真研究。在工作空间中,运用RRT-Connect算法对机械臂夹爪进行路径规划。基于Rviz、Gazebo联合仿真,采用Kinect V2视觉传感器对场景内的信息进行实时检测,传感器的点云数据传输在Rviz中可视化。通过设计一套识别抓取仿真系统,在Gazebo中对机械臂非结构环境进行仿真,完成目标的抓取,可以成功验证此方法的有效性。In a unstructured environment,aiming at the uncertainty of object recognition and grasping for manipulator,the simulation research are object recognition and manipulator motion planning in the ROS.In the work space,the RRT-Connect algo⁃rithm is used to plan the path of the grasp of the manipulator to achieve the path planning.Based on Rviz and Gazebo co-simulation,Kinect V2 sensor is used to detect the information in the scene in real time.The point cloud data of the sensor is transmitted and visu⁃alized in Rviz.By designing a grasping recognition system,and the manipulator in unstructured environment are simulated in Gaze⁃bo,the target capture is completed,the effectiveness of this method can be successfully verified.

关 键 词：机械臂 识别抓取 ROS RRT-Connect 路径规划 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]