Loading...

Design and Implementation of a Tool Ttip Force Estimation Algorithm for Surgical Robotic Systems Using Proximal Sensor using Neural Network

Mansoury, Bahman | 2020

943 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52753 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Farahmand, Farzam
  7. Abstract:
  8. One of the most important problems in the field of robotic surgery is the measurement of the laparoscopic instrument Tip force. Many efforts have been made with the use of distal sensors to issues such as high cost and need to be disinfected, but the use of proximal sensors for estimating the Tip force of the tool is very rare, which can improve distal sensor problems.In this project, first, using data obtained from simulating a set-up for the Sina robot, two static MLP, and dynamic NARX neural networks are trained to evaluate the use of machine learning algorithms to determine whether the Tip force neural network can be used. Did the robot estimate Sina's surgeon? Then, by comparing this model with the Shayeste poor model and considering the success of using the NARX neural network, then by choosing a suitable structure for the neural network model, set-up is designed and constructed to receive data from the Sina robot according to the new structure. This set includes two 6-degree-of-freedom sensors, one located at the robot arm (proximal) and the other at the distal Tip. Also in this set-up, we have to use a Spheroidal joint to attach the trocar-holder to the trocar. After mounting the set-up on the robot, the robot is given the process of data from the robot under different conditions including different forces in different directions and different longitudinal advances. At the end of the LSTM deep neural network data processing process, it is used to estimate the Tip force of the tool which can estimate the Tip force by receiving sensor data inside the arm of Sina Robot.After the feasibility stage, it was found that the NARX neural network is powered at about 0.5 N by 20 N with clean and well-controlled data. The LSTM neural network also yielded an average error of about 0.6 N in the force range of 0 to 30 N in the real situation Sina robot, which is very favorable to analytical models in this regard. It is anticipated that much better results can be achieved by modifying the data set and upgrading the neural network parameters. Also, we can use these algorithms to estimate the force applied to the grasp. Based on the results, it seems that neural networks can be used to estimate the force of surgeon robot tools proximally
  9. Keywords:
  10. Force Estimation ; Robotic Surgery ; Neural Network ; Long Short Term Memory (LSTM) ; Haptic System ; Proximal Force Sensor

 Digital Object List

 Bookmark

...see more