This project proposes a dynamic parameter identification algorithm for a 6-Degree-Of-Freedom manipulator (6-DOF) using the Recursive Least Squares (RLS) method in practice. The classical Least Squares method (LS) was commonly used to identify dynamic parameters offline. However, for time-varying systems with abrupt parameter changes, the LS method may no longer suitable to estimate dynamic parameters in real-time because the LS method is only used once all the measurements are made. To deal with this limitation, the Recursive Least Squares (RLS) method is applied in order to implement it online. To implement the RLS method, firstly, the kinematics of the system is built according to Denavit-Hartenberg (DH) notation, and the dynamic model is calculated by using Lagrange-Euler equations. After that, the dynamic equations are transformed into the general linear matrix to apply the solution of RLS method. Experiments are carried out for the 6-DOF robot with an excitaion trajectory designed for the robot. Afterwards, data acquisition is carried out by using PD controller via Simulink Deskstop Real Time, then Fourier series and low pass filter are used to analyze and process the collected signals in order to estimate dynamic parameters of the real model by using the RLS method. The estimation of the mass of the payload added to the manipualtor is also computed through the variance on some estimated inertial parameters with and without payload. Finally, the Root Mean Squared Error (RMSE) criteria is used to evaluate the accuract of the estimated results of two case studies from proposed method.

Identification Method, Least Squares estimation, Recursive Least Squares estimation, 6-DOF Robotic Manipulators, Dynamic System, Estimate Parameters

Members

POSTER