Research On Estimation Algorithm Of Kinematic Parameters Of Manipulator Based On Inertial Measurement Unit

The calibration of the kinematic parameters of the manipulator is of great significance to improve the positioning accuracy of the manipulator.Since the end effectors such as laser trackers,ballbars,and coordinate measuring instruments required by the existing calibration methods for kinematic parameters of manipulators have the disadvantages of complicated operation,large space occupation,low experimental efficiency,and high cost,so It is of great significance to find a method for calibrating the kinematic parameters of a manipulator that is easy to operate and has low experimental cost.Aiming at the problems in the existing calibration method of the kinematics parameters of the manipulator,this paper proposes a calibration method of the kinematics parameters of the manipulator based on the inertial measurement unit,that is,the inertial measurement unit as the end effector of the manipulator reads the acceleration data of the end of the manipulator in Cartesian space,the kinematics parameters of the manipulator are obtained through the normalization equation.Compared with other kinematics parameter calibration methods,the method proposed in this paper is easier to operate,more efficient,and cheaper.As the inertial measurement unit is continuously updated and upgraded,its accuracy will be higher.The kinematic parameters will be more accurate,this method has a very broad development prospect.This paper first solves the Jacobian matrix of the manipulator,and then derives the relationship between the acceleration of the end of the robot arm in Cartesian space and the angle,angular velocity,and angular acceleration in the joint space of the manipulator.After normalizing the above fomula,the kinematics parameters of the manipulator to be identified are proposed to obtain the expression of the relationship between the linear acceleration at the end of the manipulator and the kinematics parameters of the manipulator.Based on the above theory,this paper brings the kinematics model of the UR(Universal Robot)industrial robot arm into the above theory for derivation,so as to obtain the derivation formula of the UR kinematics parameter estimation algorithm used in the subsequent simulation experiments.Secondly,the recursive least squares method has the advantages of fast convergence,small memory consumption,high accuracy,and real-time calculation.Therefore,the parameter identification method adopted in this paper is the recursive least squares method.This paper first introduces the theoretical principle and derivation formula of recursive least squares method.Then,based on the above theory,the theoretical formula for parameter identification of the 6-DOF industrial robotic arm is derived.Finally,the simulation experiment of the UR10 manipulator was carried out under the ROS platform to simulate the conditions encountered in the calibration of the real manipulator kinematics parameters,and the accuracy of the final simulation results was compared and analyzed,thus illustrating the present subject based on the inertial measurement unit of the manipulator kinematics parameter identification research has certain practical application significance.