Self-data-driven Electromechanical Coupling Model Construction of Industrial Robots

As a core component of industrial robots, the health state of RV reducers directly determines the operating accuracy and response speed of industrial robots. To ensure the reliability of robot operation, it is necessary to monitor the health state of RV reducers. Generally, the built-in sensors, such as encoders and current sensors, are integrated in the motor side of the industrial robot. Therefore, it is necessary to construct an electromechanical coupled dynamics model to monitor the RV reducer with signals from the motor side. In this paper, we propose a self-data-driven dynamic modelling method to simulate the motor feedback current signal. This method constructs an electromechanical coupling model with the consideration of the internal stiffness, friction, transmission error and other nonlinear factors of the RV reducers. In addition, the influence of load variation and friction disturbance on the reducer is also considered. A self-data-driven model celebration strategy is developed to correct the model parameters automatically based on the monitoring data of the robot. An RV reducer single pendulum test bench is built to verify the proposed method. Comparison of the simulated and experimentally obtained current feedback signals is carried out in the time and frequency domains, respectively. The analysis results obtained by the model match well with the actual state response, which verifies the effectiveness of the proposed method.