An Extended Generalized Prandtl-Ishlinskii Hysteresis Model for I²RIS Robot

Area of Science:

• Robotics
• Medical Devices
• Control Systems

Background:

• Retinal surgery demands high precision due to the retina's confined anatomical spaces.
• Flexible tendon-driven robots, like the Improved Integrated Robotic Intraocular Snake (I²RIS), are developed for enhanced surgical accuracy.
• Hysteresis presents a significant challenge for precise control and positioning in these robots, particularly multi-stage hysteresis observed in small-scale systems.

Purpose of the Study:

• To develop a more accurate hysteresis model for flexible robots.
• To address the challenge of multi-stage hysteresis in the I²RIS system.
• To improve the precision and control of minimally invasive robotic surgery tools.

Main Methods:

• Proposed an Extended Generalized Prandtl-Ishlinskii (EGPI) model.
• Incorporated a novel switching mechanism into the EGPI model to describe multi-stage hysteresis.
• Validated the EGPI model using experimental data from the I²RIS robot.

Main Results:

• The EGPI model demonstrated superior fitting accuracy compared to the conventional Generalized Prandtl-Ishlinskii (GPI) model.
• EGPI outperformed GPI in terms of Root Mean Square Error (RMSE), Normalized RMSE (NRMSE), and Mean Absolute Error (MAE).
• The model accurately described multi-stage hysteresis across various motor input directions.

Conclusions:

• The EGPI model effectively models multi-stage hysteresis phenomena in flexible robots.
• This enhanced modeling capability is crucial for improving the precision of robotic-assisted minimally invasive surgeries.
• The EGPI model shows significant potential for advancing control strategies in flexible robotic systems.