Dynamic Calibration And Self-correction Of Cable-driven Parallel Robots

As a new type of robot,the cable-driven parallel robot has been plagued by its lack of accuracy.In order to improve the pose accuracy of the end moving platform,it needs to be calibrated.The paper takes the eight-cables six-degrees-of-freedom cable-driven parallel robot as the research object,and focuses on the dynamic calibration and correction for the purpose of improving the accuracy of the robot moving platform.From the aspects of robot modeling,robot moving platform pose measurement,parameter identification and error compensation,a dynamic calibration and self-correction algorithm is proposed,which aims to reduce geometric measurement and motion errors,so as to improve the moving platform pose precision.First,the kinematics model of the eight-cable six-degree-of-freedom cable-driven parallel robot is established.According to the closed principle of the vector,the nonlinear equation system of the forward solution of the moving platform pose is deduced.The Levenberg-Marquardt-quasi-Newton algorithm is used to iteratively obtain the forward solution of the moving platform pose.The real-time simulation verifies that the algorithm has good real-time performance and high precision.Then,the cable tension of the cable-driven parallel robot is analyzed and optimized.The statics and dynamics of the moving platform of the cable-driven parallel robot are analyzed,a cable tension optimization algorithm is proposed,and the practicability of the algorithm is verified by simulation,it is beneficial to reduce the pose error of the moving platform.Secondly,the kinematic error model of the cable-driven parallel robot is established,the geometric parameters in the error model are measured and analyzed,and the error parameter calibration method of the robot model is given.Error compensation and adjustment are carried out on the basis of the established error model,and a dynamic calibration and self-correction algorithm is proposed.Simulation shows that the algorithm converges and has high precision.The pose error solved by this algorithm is small,and the pose accuracy of the moving platform is improved.The real-time simulation verifies the effectiveness of the algorithm.Finally,the cable-driven parallel robot system is synthesized,an experimental prototype of eight-cable six-degree-of-freedom cable-driven parallel robot is built,and the space linear and periodic circular motion experiments are carried out respectively.The experimental results show that the proposed dynamic calibration and self-calibration strategy can make the attitude angle and position errors of the moving platform smaller,which verifies the rationality and advancement of the proposed dynamic calibration and self-calibration algorithm.