Surgical Robot Dynamic Model Identification and Control

Overview

This research develops novel

• Dynamic model identification and gravity compensation (shown) for dVRK-Si (the next-generation da Vinci Research Kit) Patient Side Manipulator (PSM) using convex optimization

• a novel force estimation method using a model-based and learning-based hybrid approach for trocar interaction.

Key Contributions

• Develop a novel force estimation approach for the robot under trocar interaction using a hybrid model combined the model-based approach and the learning-based approach, reducing the estimation errors by 30%
• Implement the dynamic model identification and gravity compensation of dVRK Classic/Si Patient Side Manipulator using convex optimization approaches, reducing the static control errors by 73%