Algorithm for identification of DC motor parameters by method of dynamic expansion of regressor and mixing

Subject of Research. The paper considers the problem of identifying the parameters of various robotic objects. A DC motor is used as an example. Existing methods for identification of parameters require either a large amount of time for accurate determination of the required values or give an estimate with a large error. We propose to expand the application area of the identification algorithm by the method of Dynamic Regressor Extension and Mixing for control problems of robotic objects with a DC motor. Method. The first stage of Dynamic Regressor Extension and Mixing method generates new regression forms by applying a dynamic operator to the original regression data. Then, the required combination of new data is selected to obtain the final desired regression form. Standard parameter estimation methods are applied to this procedure. Main Results. A new algorithm is proposed for identifying the parameters of DC motor models. It is shown that, when using the new approach, the fluctuations in parameter estimates are significantly lower, while the response time is much shorter. When using the gradient method, the transient time to estimate the signal parameters is 350 seconds, while for the Dynamic Regressor Extension and Mixing method this time does not exceed six seconds. Besides, Dynamic Regressor Extension and Mixing method has not got overshoot. Practical Relevance. The results of the work can be applied to the design of automatic control systems in control problems of electromechanical objects, including DC motors.