Novel system identification method and multi-objective-optimal multivariable disturbance observer for electric...

Area of Science:

• Robotics
• Control Systems Engineering
• Biomechanics

Background:

• Electric wheelchairs (EW) face dynamic challenges from varied terrains, slopes, and occupant weights.
• Perturbed dynamics in EWs can compromise stability and user experience.
• Accurate modeling is crucial for effective EW control system design.

Purpose of the Study:

• To develop a robust control system for electric wheelchairs (EW) that mitigates dynamic perturbations.
• To improve the stability and accuracy of EW motion across diverse conditions.
• To optimize the performance of the control system using multi-objective optimization.

Main Methods:

• Lagrange equations of motion were used to model EW dynamics, incorporating slope effects as disturbances.
• A static pre-compensator was analytically designed to decouple EW dynamics and enhance model identification.
• A two-degree-of-freedom disturbance-observer (DOB) controller was implemented for improved disturbance rejection and robustness.
• Multi-objective optimization (MOO) using NSGA-II tuned the DOB based on disturbance rejection, noise reduction, and stability.

Main Results:

• The pre-compensator effectively decoupled EW dynamics and improved model accuracy.
• The DOB controller demonstrated significant disturbance rejection capabilities.
• Experimental validation confirmed the algorithm's desirable performance and robust stability.
• The MOO approach successfully tuned the DOB for optimal control system performance.

Conclusions:

• The proposed control strategy, integrating analytical pre-compensation and DOB with MOO, significantly enhances electric wheelchair (EW) performance.
• The system exhibits robust stability and effective disturbance rejection, crucial for real-world applications.
• This research offers a pathway to more reliable and user-friendly electric mobility devices.