Fuzzy parameter identification of direct drive robots

A fuzzy identification method is proposed to estimate parameters in dynamics of direct drive robots. In this method, linear membership functions are used and the output of each fuzzy rule is assumed to be the linear combination of input variables. The input—output data for identification are generated in a recursive manner. The number and the ranges of memberships functions are determined by a heuristic searching method and a statistical training method, respectively, in the sense of least squared output errors. Once the parameters in dynamics are estimated, existing control strategies, such as the well-known computed-torque method, can be applied to make the robot follow a desired trajectory. Simulation examples are included to illustrate the validity of the proposed method.