Decentralized neural-fuzzy control of a six links robot provided with DC motors

Abstract EN

In this paper, a stable fuzzy adaptive controller for trajectory tracking is developed for a six link PUMA560 robot manipulator provided with DC motors.

The control scheme consists of two main control loops.

The first loop requires three controllers: neural-fuzzy controller I, NN controller II, and a sliding mode controller.

Neural-fuzzy controller I is a Tkagi-Sugeno fuzzy controller with adaptive neuro-fuzzy inference system (ANFIS) structure, which can improve its performance and achieves satisfactory results through online learning.

The NN controller II, is used to drive manipulator joints to compensate for the errors caused by the unstructured and structured uncertainties while controlling the robot manipulator.

The sliding mode controller, is used to make the tracking errors approach zero.

As for the second loop calculates the DC motors voltages by realizing a torque feedback.

The simulation results obtained showed that the proposed controller has achieved a high tracking performance.

In addition the results have proved the effectiveness and robustness of the proposed controller to achieve considerable enhancement of the motion tracking a robot characterized by unknown disturbances and parameter uncertainties.