It’s the season for strawberries and cream; and Wimbledon is serving up a storm in the city of London. Tasked with identifying potential injury risks, the University of Coventry has been brought on court this year to analyze muscle strain and internal loads acting on the players mid-swing.
Principal Lecturer Dr. James Shippen needed reliable, clean motion data to work with… and he used Xsens to capture it during a heated tennis rally.
How did you get brought in on the Wimbledon project?
We’d been doing some work with a tennis academy here. It’s the Edge Buster – the one that holds the pre-Wimbledon event. We were talking to those guys, explaining what we could do and they wanted to try it out on some of their players. It’s really good to work with the coaches in that they know what they want. We can measure and analyse almost anything. They wanted unusual insights; such as knee flexes, return of service, and position of the elbow – specifically how high it was being held. It was highly tailored to individual players and coaches, so we had to work very closely with them.
Did the research provide any particularly interesting or unexpected results?
I’m going to have to be careful now, because these results might be enough to identify the player in question! Again, the data was very specific to each athlete being analysed.
Overall, I suppose results defied expectation. One coach was there when we were grabbing the data, and he thought the player was moving as instructed. That’s what it looked like. But then when we actually looked at the data, we found the player was doing something completely different. It shows the value of getting objective information rather than just eyeballing a movement.
Can you describe the process of capturing data for Wimbledon athletes?
We were putting athletes into the Xsens MVN Link suit. Coaches would then carry out an average training session while we were recording the data. Partway through, the coach would explain what he was doing and the motions he was interested in. We would then bring up the data, feed it into our custom developed Biomechanics of Bodies (BoB) software, and supply him with the information required. If we want kinematic information from a rally, that processes in about 2-3 minutes. It can all be done right on the court. We plug it through and there it is.
We wanted to be the silent partner, as it were, so the lesson could continue as usual. Xsens allowed us to do that. The MVN Link brought us reliable data, unobtrusively, in real time. And clean data, at that.
We’re used to handling the Xsens system so we can get reliable data very quickly. And we trust the data that comes through. We’ve tested it against control groups before.
What features of Xsens make it well-suited for sports and biomechanics?
You could not have done what we did using an optical setup, because we’re outside, actually on the tennis court. We’re getting clean data, in real-time, with inertial suits. An optical solution would not allow capture in the true-to-life environments achieved with Xsens MVN Link.
With the Xsens system, we can gather data across the whole tennis court, regardless of extraneous activities. We actually used it to successfully capture two tennis players simultaneously. Again, it would be very difficult to do that with an optical stage.
For us, Xsens was the ideal system in terms of portability. Just getting it to the tennis court was easy as can be. The setup time is reasonable. The coverage is wonderful. We almost just take the data as read now – it’s that good.
What was your experience like, capturing two players at once?
We had certain issues synchronising the two captures at first, as we were running Xsens on two completely different laptops. Before the start of a rally, we attempted to synchronise the laptops as much as possible, then spent a while eyeballing the time shift to pull players back to a baseline. Then we could place the two avatars into the same ‘digital tennis court’. The two Xsens systems didn’t interfere with each other at all and we got good data; insights we couldn’t have achieved through any other method.
How will this data eventually assist athletes in terms of injury prevention or performance?
We see it in two different ways. First, if an athlete is playing normally – at peak fitness and doing as well as they can – we can measure and keep a record of that movement pattern. Then if the athlete incurs an injury at any point in the future, we can refer back to their records. For instance, physiotherapists could see how far along in recovery an athlete’s shoulder might be, and how long it will likely take to get back that previous range of motion.
In terms of injuries, you could feed this information into our BoB software. Short for ‘Biomechanics of Bodies’, it can automatically analyse internal body loadings, joint contact forces, muscle strain, and more. It’s absolutely excellent – and we don’t just say that because we built it! From BoB, you can work out whether more energy is being demanded from a muscle than it can generate.
You can also use Xsens to compare athletes’ various levels of performance. We might take Novak Djokovic and compare him to a local county player. We can then accurately tell what the difference is. That’s optimisation of performance.
What’s in store for Coventry University and Xsens in the future?
We’re hoping Wimbledon will be a stage from which this type of analysis can take off, using Xsens for more and more tennis-focused projects. It’s an emerging technology, and the tennis world is very much built on tradition, so adoption will take time. If biomechanics and tennis can just slowly move forwards together, then we’ll be happy.
The next major event would be the US Open at the tail end of August. We’d like to get some presence there, if possible. We’ve also got a lot of interest from the Middle East.
If you want reliable, clean data – that you know is going to work – get an Xsens system.
Xsens Motion Capture Whitepaper
Xsens MVN Analyze : Consistent Tracking of Human Motion Using Inertial Sensing
This whitepaper describes key characteristics and shows an analysis of the performance of the new engine. The performance analysis includes a comparison with an optical position measurement system in combination with OpenSim for walking data, as well as a consistency analysis for running data.