Design Of Mobile Manipulator System Based On Human-machine Shared Control

Mobile manipulator is an important form of robots,which are widely used in life services,industrial manufacturing,and space exploration.The existing control methods of mobile manipulators mainly include teleoperation and robot autonomous control.The operation mode of the teleoperation robot system is simple,but factors such as long-distance communication delay and the ability of the operator seriously affect the control accuracy and stability of the system;the autonomous control of the robot has high control accuracy and can replace people to complete some simple and fixed tasks,But based on the existing control methods and sensor technology,it is difficult to develop a completely autonomous control robot system for complex environments and tasks.Therefore,fusing the precise control of the robot with the judgment and decisionmaking of the human to form a shared human-machine control is an effective method to solve the above-mentioned problems.The thesis studies a mobile manipulator system based on human-machine shared control.The main work is summarized as follows:1.The overall design of the mobile manipulator human-machine sharing control system.In order to make the system suitable for complex environments and perform diversified tasks,the shared control is applied to the mobile manipulator system,the structure of the human-machine shared control system composed of the user end and the robot end is designed and the operating principle of the system is analyzed.2.The human-machine sharing control strategy of the mobile platform is designed.The mobile platform adopts a shared control strategy based on human-machine mutual trust and environmental information.First,it analyzes the autonomous path planning method of the mobile platform;then integrates the operator status,robot performance and environmental information to establish a human-machine mutual trust model;finally adopts a fuzzy control method to human-machine mutual trust and environmental information are used as inputs,and the shared control coefficients are reasonably allocated based on the expert experience in the actual work of the robot.3.The human-machine shared control strategy of the manipulator is designed.The manipulator adopts a shared control strategy based on operator control and robot vision guidance.The kinematics of the robotic arm is first analyzed;then the manipulator system target recognition and positioning,and autonomous path planning methods are described;then the reach of the manipulator,the detection depth of the camera,and the The ability of the operator,the working space is divided into the operator's hand control area,the manipulator visual guidance area and the non-grabable area;finally,the control authority is switched according to the working area where the end of the robotic arm is located.4.System integration and experimental analysis.Based on the Robot Operating System(ROS),the system navigation and robotic arm grasping function modules are developed.Design the robot teleoperation,autonomous control and shared control experiments.The experimental verification is carried out on the stage simulation platform and the Turtlebot2 i physical platform built.The results show that the mobile manipulator system based on human-machine shared control proposed in this thesis has higher safety,Control accuracy and interactivity.