Callum Barnes, a physics PhD student at Kent, has been on an incredible journey to reach a career highlight of working with Norwegian triathlete and world champion Kristian Blummenfelt at this year’s Paris Olympics and World Ironman Championship.
Callum completed his Physics with a Foundation Year BSc at Kent before gaining experience with Santander UK via the SEPnet internship programme. His talent in using Machine Learning techniques in his academic work drove him to pursue his research interests further in an area he was passionate about – cycling. So, after an internship with UK start up Body Rocket, which developed the world’s first on-bike system to measure the drag force of cyclists, he began an industrial partnership with the company alongside his PhD. He is now Head of AI in Aerodynamics at Body Rocket, while he investigates the application of machine learning techniques in cycling – his overarching PhD enquiry.
Here Callum tells us how his PhD work has been feeding into his career and how his PhD has made a difference to his development.
What have been the focus research interests of your PhD?
In the first year of my PhD I investigated the impact of shuffling on the saddle. Shuffling is a well-known phenomenon within the cycling fraternity and is the action of sliding forward onto the tip of the saddle, when aggressively cycling, then sliding back.
In my second year, I have gone on to research the application of machine learning techniques on Body Rocket system data to determine the position of a rider on the bike. I also investigated the application of computer vision techniques to determine the joint location of riders on a bike with the aim to tie this into the Body Rocket system data.
I am now working on the creation of an artificial intelligence chatbot that can interpret the proprietary Body Rocket data and guide a rider through the process of getting more aerodynamic. The aim of this chatbot is to interpret this data, learn an individual, and tell them how to get faster, with no guessing, no false information, just facts, knowledge and understanding tailored to you. I’m aiming to create an ‘aerodynamics coach’ which will give complete novices access to aerodynamics knowledge tailored to them.
What inspired you to explore these research areas?
I wanted to make it possible for a rider to find their optimal position and track it in real time, while riding. Using the rich dataset produced by the Body Rocket system and machine learning, I have managed to do this while also giving riders the ability to see how well they were maintaining their position and to adapt depending on conditions.
Artificial Intelligence (AI) is constantly innovating, how does this impact your PhD research?
With the field of AI as it is I need to be quick to adapt to ensure that my research is current and relevant. Recently I attended the first Conference on Language Modelling to gain a better understanding of the state of the art in this fast-paced field. This was an international conference hosted in Philadelphia, USA with academics from Google, Meta and top US institutions. This allowed me to gain a deeper understanding of not only the field but current techniques and applications of Language Modelling.
What is the benefit of an industrial partnership alongside your PhD?
An industrially sponsored PhD allows me to network with not only academics, but professionals in the industry. Due to the nature of my PhD and partnership I have attended conferences, expositions, sales events, Investor events and data acquisition trips supporting athletes.
How has your PhD benefited your career so far?
The first testing of my AI chatbot was on the university’s supercomputer and recently I was in the on-site motion capture facilities collecting data for another research project. The benefits don’t stop at the facilities either. I am a founding member of the Natural Sciences Machine Learning Network where we hold monthly research meetings for all academics in the Natural Sciences that want to involve machine learning and AI in their research.
It has also helped to get involved in research projects outside of my PhD. For example, I have developed an automated system for detecting doping agents within urine samples for a lab in Utah, and on the horizon is some work in collaboration with the NHS to do some similar work which could even help detect early signs of kidney transplant rejection and Alzheimer’s.
My PhD is allowing me to make real world impact.
How has your PhD been feeding into the development of Body Rocket products?
My research has allowed me to provide value to Body Rocket via my research, such as research on positional identification, shuffling and my artificial intelligence chatbot.
My PhD research has helped to make the Body Rocket system even easier to use, as a result it is a key component of the Body Rocket system as a product.
How has it been working with professional triathletes like Kristian Blummenfelt and Gustav Iden on their cycling performances?
It has been an amazing experience, knowing I am able to make a positive impact through my research and work with Body Rocket. It has been invaluable to be supporting athletes at the elite level and it is reassuring evidence to prove the value of my research in the real world. I recall watching the Olympic triathlon on Paris 2024 with eager eyes as I helped with the set-up that Kristian was riding, backed by data and science. More recently after a trip in Flagstaff, Arizona, USA I was working with Gustav and Kristian for their preparations for the World Ironman Championships in Kona. The positions that I helped set them up in created a media frenzy in the triathlon and cycling world, who were obsessed with how crazy their positions were.
What advice would you give to anyone thinking about doing a PhD and working in the sports industry at the same time?
Attend events and become well versed in the current state of the sport.
