Positioning Error Calibration of Industrial Robots Based on Random Forest

Because most industrial robots are taught using the direct teaching and playback method, they are unsuitable for variable production systems. Alternatively, the offline teaching method has limited applications because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have been conducted to calibrate the position and posture. Positioning errors of robots can be divided into kinematic and non-kinematic errors. In some studies, kinematic errors are calibrated by kinematic models, and non-kinematic errors are calibrated by neural networks. However, the factor of the positioning errors has not been identified because the neural network is a black box. In another machine learning method, a random forest is constructed from decision trees, and its structure can be visualized. Therefore, we used a random forest method to construct a calibration model for the positioning errors and to identify the positioning error factors. The proposed calibration method is based on a simulation of many candidate points centered on the target point. A large industrial robot was used, and the 3D coordinates of the end-effector were obtained using a laser tracker. The model predicted the positioning error from end-effector coordinates, joint angles, and joint torques using the random forest method. As a result, the positioning error was predicted with a high accuracy. The random forest analysis showed that joint 2 was the primary factor of the X- and Z-axis errors. This suggests that the air cylinder used as an auxiliary to the servo motor of joint 2, which is unique to large industrial robots, is the error factor. With the proposed calibration, the positioning error norm was reduced at all points.

A Novel Error Equivalence Model on the Kinematic Error of the Linear Axis of High-End Machine Tool

The kinematic errors of the linear axis play a key role in machining precision of high-end CNC (Computer Numerical Control) machine tool. The quantification of error relationship is still an urgent problem to be solved in the assembly process of the linear axis, especially considering the effect of the elastic deformation of rollers. A systematic error equivalence model of slider is proposed to improve the prediction accuracy for kinematic errors of the linear axis which contains the base, the linear guide rail and carriage. Firstly, the geometric errors of assembly surface of linear guide rail are represented by small displacement torsor. According to the theory of different motion of robots, the error equivalence model of a single slider is established, namely the geometric error of assembly surface of linear guide rail and the pose error of slider is equivalent to the elastic deformation of roller. Based on the principle of vector summation, the kinematic error of a single slider is mapped to the carriage and the kinematic error of the linear axis is obtained. Besides, experiments validation of kinematic error model of the linear axis is carried out. It is indicated that the proposed model is accurate and feasible. The proposed model can provide an accurate guidance for the manufacturing and operation performance of the linear axis in quantification, and a more effective reference for the engineers at the design and assembly stage.

Absolute Positioning Accuracy Improvement in an Industrial Robot

The absolute positioning accuracy of a robot is an important specification that determines its performance, but it is affected by several error sources. Typical calibration methods only consider kinematic errors and neglect complex non-kinematic errors, thus limiting the absolute positioning accuracy. To further improve the absolute positioning accuracy, we propose an artificial neural network optimized by the differential evolution algorithm. Specifically, the structure and parameters of the network are iteratively updated by differential evolution to improve both accuracy and efficiency. Then, the absolute positioning deviation caused by kinematic and non-kinematic errors is compensated using the trained network. To verify the performance of the proposed network, the simulations and experiments are conducted using a six-degree-of-freedom robot and a laser tracker. The robot average positioning accuracy improved from 0.8497 mm before calibration to 0.0490 mm. The results demonstrate the substantial improvement in the absolute positioning accuracy achieved by the proposed network on an industrial robot.

Error estimation and sensitivity analysis of the 3-UPU translational parallel robot due to design parameter uncertainties

This work deals with the estimation and the sensitivity analysis of the 3-UPU parallel robot error. Based on the Newton–Euler formalism, the robot dynamic model is given in a closed form. This model is validated by the software ADAMS. Using the interval analysis method, a new algorithm is proposed, which estimates the errors in the motion of the end-effector and the errors in the actuator forces as a function of the design parameters uncertainties. The obtained results show that the kinematic errors are minimal at the workspace center. Moreover, these errors increase as the platform moves along the vertical axis. It is also shown that kinematic errors in the actuator joints are the most influential parameters on the manipulator accuracy. Therefore, using actuators with a higher accuracy can highly reduce the errors in motion of the platform.