A validation procedure to identify joint friction, reductor self-locking and gear backlash parameters.

The paper presents a new method for the identification of joint friction, reductor self-locking and gear backlash parameters. The method was developed using an existing hexapod robot. During the process, the motor driving currents given by the model and measured on the real device were observed and compared. Initially, the current curves showed specific differences. After the analysis of these deviations, their causes were identified as joint friction, reductor self-locking and gear backlash. The parameters of these phenomena were determined using swarm optimization. As a result of this process, a validated model was obtained. The mentioned mechanical properties were identified using a step-by-step method, in which one property aided in the discovery of another. The robot model was created in a MATLAB/Simulink environment, using the Simscape Multibody Toolbox. [ABSTRACT FROM AUTHOR]