An investigation of error compensation for a 6-DoF industrial robot based on neural network and stiffness modelling

Abstract

With the development of intelligent manufacturing, the role of industrial robots is becoming more and more important. However, the relatively low absolute positioning accuracy limits industrial robot application in high precision manufacturing. The main reason for the low positioning accuracy of industrial robots comes from the series configuration and insufficient stiffness, which leads to large motion errors. This paper proposed an error compensation method based on BP neural network combined with industrial robot stiffness model. Firstly, the relationship between the joint angles, the space stiffness and the error of the industrial robot is established through the stiffness model. Then, the neural network training set was constructed based on the experimental data and the simulation data from the established stiffness model. Finally, based on the training results of BP neural network, the spatial positioning error of the 6-DOF industrial robot was measured and compensated. Experimental results show that the error compensation method based on BP neural network increases the position accuracy by 95%, and the spatial position error is reduced to less than 0.005mm. This validates that the working performance and accuracy of the industrial robot can be improved, which is helpful for the further application of industrial robot in precision machining and measurement.