Needle target-insertion trajectory planning based on reforcement learning expert's skill

Authors:
Bi Dexue;Li. Zeguo;Xue Qiang;Yu Demin
Affiliations:
Mechanical Engineering Department, Tianjin University of Scienc and Technology, Tianjin, P.R.China;Mechanical Engineering Department, Tianjin University of Scienc and Technology, Tianjin, P.R.China;Mechanical Engineering Department, Tianjin University of Scienc and Technology, Tianjin, P.R.China;-
Venue:
ROBIO'09 Proceedings of the 2009 international conference on Robotics and biomimetics
Year:
2009

Citing 3
Cited 0

ArtDefo: accurate real time deformable objects

Proceedings of the 26th annual conference on Computer graphics and interactive techniques
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping

ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

this paper proposes a new robot needle insertion trajectory planning method based on learning expert's skill. Through reforcement learning, the system can imitate the expert's behavior in planning optimal needle insertion policy. After learning two experts' skill and experience, the needle insertion optimal policy shows that each one can catch the main characters of the expert's own behavior. Through experimental verification, this paper also presents an approach on improving system learning speed. This makes it possible for robot needle trajectory real time enforcement learning and target insertion in complicate surgical operating conditions.