High precision control device and method for industrial robot

Abstract

The present invention proposes a high-precision control device and method for industrial robots, including: using a six-dimensional force-moment sensor installed between the base of the industrial robot and the fixed surface to measure the transmission of each connecting rod during the operation of the industrial robot to Support reaction force and moment of the base, and observe the joint motion dynamics; conduct online real-time estimation and correction on the observed linear dynamic model parameters, and obtain the linear model estimated joint motion dynamics τ d,model ; Use the joint drive torque τ motor and a linear model to estimate the joint motion dynamics τ d,model , online estimation of joint friction τ friction ; According to the parameters obtained from joint motion dynamic observation, online estimation of joint friction force τ friction , using a model-based robot control algorithm to achieve high-precision control of industrial robots. The invention realizes the decoupling identification of the friction parameter and the robot mass distribution characteristic parameter, obtains a relatively accurate robot dynamics model online, and can be applied to scenes requiring high control precision of industrial robots.

Description

technical field [0001] The invention relates to the technical field of industrial robots, in particular to a high-precision control device and method for industrial robots. Background technique [0002] Model-based control is an important technical approach to achieve high-precision control of industrial robots, but the premise of this method is to obtain a sufficiently accurate robot dynamic model, including the mass distribution characteristic parameters of the robot structure, friction parameters of the transmission system, etc. At present, there are two main ways to obtain the dynamic parameters of industrial robots: the nominal model parameters given by design and empirical parameters and the robot dynamic parameters identified by the robot body parameter identification technology. The problem with the first method is that due to the processing Due to the existence of production deviation, the actual kinetic parameters are quite different from the nominal parameters, an...

AI Technical Summary

Problems solved by technology

At present, there are two main ways to obtain the dynamic parameters of industrial robots: the nominal model parameters given by design and empirical parameters, and the robot dynamic parameters identified by the robot body parameter identification technology. The problem with the first method is that due to the processing Due to the existence of production deviation, there is a big difference between the actual kinetic parameters and the nominal parameters, and the application of nominal parameters cannot accurately describe the actual kinetic characteristics

The second method can obtain dynamic model parameters with a certain accuracy. It is widely used at present, but there are certain problems. For example, the parameter identification is carried out in the same group of motions, and the inertial force, Coriolis force, gravity and friction force in the driving torque are different. The specific gravity is determined by the characteristics of the identification motion. In order to fully identify the components, the identification motion needs to be carefully designed. In addition, the friction force is affected by the actual working conditions. The working conditions are not exactly the same, so the dynamic model parameters obtained by the above two methods are difficult to meet the requirements of model-based control of high-precision industrial robots

[0003] In some model-based industrial robot control schemes, in order to eliminate the uncertainty of the dynamic model parameters, the adaptive control method is used to modify the model parameters online, which has a certain effect, but the problem is that the friction and motion force parameters in the joint torque Estimated during the measurement, due to the influence of Coulomb friction nonlinearity, the convergence accuracy of the online identification of dynamic parameters is limited

Images