Related Experiment Video
Updated: Nov 5, 2025
13:44
Haptic/Graphic Rehabilitation: Integrating a Robot into a Virtual Environment Library and Applying it to Stroke Therapy
Published on: August 8, 2011
14.2K
A hybrid learning-based hysteresis compensation strategy for surgical robots
Qian Gao1,2, Ning Tan3, Zhenglong Sun1,2
1School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China.
Summary
This study presents a new gravity compensation strategy for the da Vinci Research Kit's Master Tool Manipulator (MTM) that effectively separates and models cable hysteretic forces, improving accuracy and reducing data needs.
Area of Science:
- Robotics
- Control Systems
- Machine Learning
Background:
- Hysteretic forces from external electric cables disturb precise gravity compensation in the da Vinci Research Kit's Master Tool Manipulator (MTM).
- These nonlinear disturbance forces often have the same magnitude as gravitational forces, complicating parameter estimation.
Purpose of the Study:
- To develop a novel strategy for separating and modeling hybrid hysteretic and gravitational forces acting on the MTM.
- To enhance the precision of gravity compensation models for robotic manipulators.
Main Methods:
- A learning-based approach was employed to model the distinct forces.
- A specialized Elastic Hysteresis Neural Network was utilized to accurately capture the nonlinear hysteresis of disturbance forces.
Main Results:
- The proposed strategy achieved higher compensation accuracy, ranging from 78.64% to 93.32%.
- The method required fewer real-world samples for model estimation, needing only 100 samples per joint.
Conclusions:
- The developed gravity compensation strategy significantly improves performance compared to existing methods for the MTM.
- Experimental validation confirms the strategy's effectiveness and superiority in comparative analyses.
