Neural Network-Based Cooperative Trajectory Tracking Control for a Mobile Dual Flexible Manipulator

Area of Science:

• Robotics
• Control Systems Engineering
• Artificial Intelligence

Background:

• Mobile dual flexible manipulator (MDFM) systems present challenges in cooperative control due to unknown dynamics and time-varying trajectories.
• Accurate trajectory tracking and vibration suppression are critical for the effective operation of such robotic systems.

Purpose of the Study:

• To develop a robust cooperative trajectory tracking control (CTTC) strategy for MDFM systems operating under uncertain dynamics.
• To address the challenge of tracking time-varying trajectories while simultaneously suppressing system vibrations.

Main Methods:

• Establishment of the dynamic model for the wheeled mobile manipulator system in a 2-D space.
• Approximation of unknown system dynamics using a radial basis function neural network (RBFNN) structure.
• Design of a CTTC strategy incorporating a servo system for cooperative operation and vibration mitigation.

Main Results:

• The proposed RBFNN-based CTTC strategy effectively handles unknown system dynamics.
• The control scheme enables precise tracking of time-varying trajectories for the MDFM system.
• Numerical simulations and theoretical analysis confirm the successful vibration suppression and cooperative operation.

Conclusions:

• The developed CTTC strategy provides a robust solution for cooperative trajectory tracking in MDFM systems with unknown dynamics.
• The integration of RBFNN and servo systems offers a promising approach for enhancing the performance and stability of flexible robotic manipulators.
• This research contributes to advancing control methodologies for complex robotic systems in dynamic environments.