Explicit Model Predictive Control of a Magnetic Flexible Endoscope.

Area of Science:

• Robotics and Control Systems
• Biomedical Engineering
• Magnetic Actuation

Background:

• Flexible endoscopes are crucial for minimally invasive diagnostics.
• Precise control of magnetically actuated devices remains a challenge.
• Existing methods lack advanced closed-loop control for magnetic capsules.

Purpose of the Study:

• To develop and implement an explicit model predictive control (EMPC) strategy for a magnetically actuated flexible endoscope.
• To compute the precise motion of an external permanent magnet for desired forces and torques.
• To address the nonlinear dynamics and constraints of magnetic actuation systems.

Main Methods:

• Development of a simplified dynamic model using Euler-Lagrange equations.
• Application of explicit model predictive control (EMPC).
• Novel single-step nonlinear optimization for external magnet positioning.
• Implementation on an experimental platform with a realistic colon phantom.

Main Results:

• Successful computation of external magnet motion based on desired forces and torques.
• Demonstration of workspace boundary and actuation constraint consideration.
• Validation of the control strategy's effectiveness through bench-top trials on a colon phantom.
• First reported instance of advanced closed-loop control for magnetic capsules.

Conclusions:

• The proposed EMPC strategy effectively controls magnetically actuated flexible endoscopes.
• The technique shows promise for various magnetically manipulated payloads, including endoscopes and wireless capsules.
• This work pioneers model-based control for magnetic capsule manipulation.