| The introduction of the intelligent manufacturing 2025 strategy and the emergence of China's aging problem have brought technical support and application space to dexterous operation robots used in the home and service fields.The dexterous operation robot mainly uses the organic combination of omnidirectional mobile chassis,robot manipulator,pan/tilt,lifting mechanism and various sensors,and is equipped with intelligent algorithms for autonomous movement,target recognition and intelligent grasping,providing water supply,lighting and other services for the elderly and disabled people to assist them in their daily life and work.This paper studies the dexterous operation robot system and its calibration methods.The main research contents are as follows:Firstly,the design of the dexterous robot system,including robot structure design,control system design(software,hardware)and auxiliary system design.The robot structure design mainly draw up the design plan and parameters according to the functional requirements,and combines the processing technology and strength analysis to design the parts;The control system design is divided into two parts: software and hardware.The hardware part of the control system mainly introduces the system composition and control scheme;The software framework of the robot and the manipulator is outlined in part,and the auxiliary system of the robot is designed for the display module and the power module part.Secondly,the kinematics and workspace analysis of the robot including the mathematical modelling of omnidirectional mobile chassis kinematics,provides a theoretical basis for the precise control of the robot;the forward and inverse kinematics analysis and workspace analysis of the four-axis manipulator;The omnidirectional mobile chassis,the secondary lifting and the four-degree-of-freedom manipulator are regarded as Stanford-like arm manipulator,and the DH parameter table is established.The Monte Carlo method is used to solve the workspace through Matlab.And in the virtual simulation software(V-REP),the simulation verification of the robot's working space provides a theoretical basis for subsequent control and grabbing.Then,based on the dexterous operation robot experimental prototype,the error calibration experiment of the robot odometry and the manipulator is performed.Based on the kinematic parameter correction of omnidirectional mobile robot,the odometry calibration of the robot is realized.The recursive formula of the robot pose is obtained according to the kinematics model,and the expression of the estimated value of the kinematic parameters is obtained based on the least squares method of multiple linear equations.During the experiment,the pose information of the robot is obtained through the visual sensor.After several experiments,the kinematic parameter estimation value is obtained,and the experimentally verified,the calibrated robot odometry information is more accurate;The D-H parameter of the manipulator is calibrated based on the axis measurement method.The manipulator is uniaxially moved,and the position of the end of the manipulator is obtained based on the visual measurement method.Then the identified D-H parameters are obtained by spatial circle fitting and least squares method,and the identification result is written into the controller.The initial point is regarded as the reference,the arm moved before and after calibration,the difference between the end position and the reference is compared to verify the calibration effectiveness.The analysis of the standard deviation,the mean value and the maximum value proves that the positioning accuracy after calibration is more accurate.Finally,the navigation of the mobile robot,the trajectory planning of the manipulator and the experiment of the robot are carried out,and the coordination of the multi-sensor and multiple subsystems is combined to verify the function of the robot.In the experiment,the stability of the overall structure of the robot,the accuracy of the control system,and the intelligence of the navigation and planning algorithms were verified by setting different scenes,different tasks and different disturbances. |
