Fuzzy Neural Network Control of a Flexible Robotic Manipulator Using Assumed Mode Method

Area of Science:

• Robotics and Control Systems
• Mechanical Engineering
• Artificial Intelligence

Background:

• Flexible structures present challenges in precise motion control due to inherent vibrations.
• Traditional control methods often struggle to simultaneously achieve accurate trajectory tracking and effective vibration suppression.

Purpose of the Study:

• To develop and validate a novel fuzzy neural network (NN) control strategy for single-link flexible structures.
• To accurately track desired trajectories while maximally suppressing vibrations.
• To rigorously ensure system stability using Lyapunov's method.

Main Methods:

• Assumed mode method for developing the discrete dynamic model of the single-link flexible structure.
• Fuzzy neural network (NN) control design for trajectory tracking and vibration suppression.
• Lyapunov's stability method for proving uniform ultimate boundedness.

Main Results:

• The fuzzy NN control strategy effectively tracks desired trajectories.
• Maximal suppression of flexible vibrations was achieved.
• Lyapunov's stability analysis confirmed uniform ultimate boundedness of the system.
• Simulations and experiments on the Quanser platform demonstrated the proposed control's effectiveness and feasibility.

Conclusions:

• The proposed fuzzy NN control is a viable and effective strategy for controlling single-link flexible structures.
• It offers superior performance in trajectory tracking and vibration suppression compared to proportion derivative control.
• The method provides rigorous stability guarantees.