Adaptive Kinematic Modelling for Multiobjective Control of a Redundant Surgical Robotic Tool

Francesco Cursi,Petar Kormushev,George P Mylonas

doi:10.3390/robotics9030068

Abstract

Accurate kinematic models are essential for effective control of surgical robots. For tendon driven robots, which are common for minimally invasive surgery, the high nonlinearities in the transmission make modelling complex. Machine learning techniques are a preferred approach to tackle this problem. However, surgical environments are rarely structured, due to organs being very soft and deformable, and unpredictable, for instance, because of fluids in the system, wear and break of the tendons that lead to changes of the system’s behaviour. Therefore, the model needs to quickly adapt. In this work, we propose a method to learn the kinematic model of a redundant surgical robot and control it to perform surgical tasks both autonomously and in teleoperation. The approach employs Feedforward Artificial Neural Networks (ANN) for building the kinematic model of the robot offline, and an online adaptive strategy in order to allow the system to conform to the changing environment. To prove the capabilities of the method, a comparison with a simple feedback controller for autonomous tracking is carried out. Simulation results show that the proposed method is capable of achieving very small tracking errors, even when unpredicted changes in the system occur, such as broken joints. The method proved effective also in guaranteeing accurate tracking in teleoperation.

Highlights

Accuracy and precision are of utmost importance in many robotic applications, especially in minimally invasive surgery, where little damage should be inflicted on the patient’s body
Because of safety issues related to the accuracy and capabilities of surgical robots, and ethical issues due to patients not feeling comfortable being operated by autonomous robots, most of the surgical robotic platform are still teleoperated: the surgeon directly controls the surgical robot through a master device
We presented an approach to control a redundant surgical robot, consisting in learning the forward kinematic model of the robot by means of a global learning method such as Artificial Neural Networks (ANN)

Summary

Introduction

Accuracy and precision are of utmost importance in many robotic applications, especially in minimally invasive surgery, where little (or preferably no) damage should be inflicted on the patient’s body. Because of safety issues related to the accuracy and capabilities of surgical robots, and ethical issues due to patients not feeling comfortable being operated by autonomous robots, most of the surgical robotic platform are still teleoperated: the surgeon directly controls the surgical robot through a master device. This means that the precision of the robotic system is still not fully exploited, because of the possible motion errors coming from the surgeon. Autonomous surgery, instead, allows faster and more precise execution, and reduction of the surgeon’s burden [1]

Methods

Results

Discussion

Conclusion