In this section

Human-Centred Automation, Robotics and Monitoring

We use perceptual methods, AI, and frugal robotics innovation to deliver transformative diagnostic and treatment solutions.

Head of Group

Dr George Mylonas

B415B Bessemer Building
South Kensington Campus

+44 (0)20 3312 5145

YouTube ⇒ HARMS Lab

What we do

The HARMS lab leverages perceptually enabled methodologies, artificial intelligence, and frugal innovation in robotics (such as soft surgical robots) to deliver transformative solutions for diagnosis and treatment. Our research is driven by both problem-solving and curiosity, aiming to build a comprehensive understanding of the actions, interactions, and reactions occurring in the operating room. We focus on using robotic technologies to facilitate procedures that are not yet widely adopted, particularly in endoluminal surgery, such as advanced treatments for gastrointestinal cancer.

Meet the team

Mr Amirhosein Alian
Research Postgraduate

Dr James P Avery
Honorary Lecturer

Mr Junhong Chen
Research Postgraduate

Mr Kaizhong Deng
Research Postgraduate

Mr Hanqi LOU
Research Postgraduate

Dr George Mylonas BEng, MSc, DIC, PhD
Senior Lecturer in Robotics and Technology in Cancer

Dr Adrian Rubio Solis
Research Associate in Sensing and Machine Learning

Dr Jianlin Yang
Research Associate

Mr Zhangxi Zhou
Research Postgraduate

Citation

BibTex format

@article{Cursi:2020:10.3390/robotics9030068,
author = {Cursi, F and Mylonas, GP and Kormushev, P},
doi = {10.3390/robotics9030068},
journal = {Robotics},
pages = {68--68},
title = {Adaptive kinematic modelling for multiobjective control of a redundant surgical robotic tool},
url = {http://dx.doi.org/10.3390/robotics9030068},
volume = {9},
year = {2020}
}

Download

RIS format (EndNote, RefMan)

TY - JOUR
AB - 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.
AU - Cursi,F
AU - Mylonas,GP
AU - Kormushev,P
DO - 10.3390/robotics9030068
EP - 68
PY - 2020///
SN - 2218-6581
SP - 68
TI - Adaptive kinematic modelling for multiobjective control of a redundant surgical robotic tool
T2 - Robotics
UR - http://dx.doi.org/10.3390/robotics9030068
UR - https://www.mdpi.com/2218-6581/9/3/68
UR - http://hdl.handle.net/10044/1/82671
VL - 9
ER -