Why I chose the Robotics CDT

Writes Jonathan Bowles

Hello and welcome to the Edinburgh Centre for Robotics student blog! My name’s Jonathan and I’m one of the first-year (Masters level) students in the centre. I want to talk to you about my experiences of the programme so far, my research project and why I think Edinburgh is a great place to study if you are interested in research in Robotics.

I started the programme quite recently, in September, with 16 other students from all around the world. The Centre for Doctoral Training (CDT) is structured quite differently from the traditional approach to studying for a Ph.D. – students are recruited as part of a ‘cohort’ and the first year acts as (very intensive) training for the topic which you will pursue as part of your Ph.D. – I am currently studying machine learning, kinematics, robot navigation, control theory and computer vision. The students have come from a huge range of different backgrounds – from all over the world and from lots of different branches of science and engineering – my own background is in electrical & electronic engineering, but there are also computer scientists, mechanical engineers, physicists and more. I think it makes for quite an interesting mix – the benefits are very apparent when working in laboratories, where different students can contribute in different ways. It’s also a reflection on the field of Robotics itself – making something as complex as, say, a humanoid robot requires a very broad range of different skills.

Some students have yet to decide on the final topic for their Ph.D., whereas others have applied for industrial studentships – there are already some vacancies open for next year. My own project – ‘Robust Sensory Interfaces for Advanced Prosthetics’ – is sponsored by a Scottish manufacturer of cutting-edge myoelectric prosthetics, Touch Bionics. Touch Bionics are the makers of the i-limb – a prosthetic arm which the user can control through an electromyography (EMG) sensor interface. When the user contracts a particular muscle site, the sensors pick up the electrical activity in the muscles, which is then decoded and translated into motion.

I-limb-coin

The problem is that EMG sensors are quite a tricky technology to work with – small movements in the placement of the sensor can lead to problems decoding the EMG signals. As it is not practical to implant sensors under the skin (where they would pick up ‘cleaner’ signals) on a long-term basis, the company is exploring other sensor options, which is where the project comes in. I will be working with Professor Sethu Vijayakumar at the University of Edinburgh and Dr Ravinder Dahiya at the University of Glasgow on designing a new sensor suite for the company.

The project cuts across lots of different scientific disciplines – machine learning, sensor fabrication, signal processing, control theory and more – which I think are very interesting in their own right, so it’s a great privilege to be able to do research that touches on them all. This inter-disciplinary approach is one of the great things about the programme, and one of the reasons why I would definitely encourage students who are interested in research in Robotics to apply – you could design part of a system which is going to transform the lives of people around the world.


 

If you wish to contact me, simply email enquiries@edinburgh-robotics.org

Advertisements