Developing IMU Sensors For Capturing Motion In Sports

IMU sensors are pretty useful because when strapped to the right location and given the right context they can provide very insightful information about an athlete’s (or anyone’s) movements. In this post, we are going to look at a couple of options in the market that allows us to skip the hardware development and jump right into the application development. Feel free to skip to the different sections that interest you:

[ Intro To IMUsmbientlabsNotch SensorNotch Mocap TestCustom Sensors]

Intro To IMUs

In case this is the first time you are hearing about IMU, here’s a brief intro. IMU stands for Inertial Measurement Unit; it is an electronic device that typically has accelerometers, gyroscopes and magnetometers, and it measures its own acceleration, angular rate (or spin rate) and surrounding magnetic field. IMUs are not only used in sports, in fact, it is used in many consumer electronic devices. Our smartphones for one has IMUs for detecting the orientation of the phone and changing the display to portrait or landscape. The IMUs also allows for functions such as undoing texting errors, a spirit level and motion sensor games. If a user carries the phone with them in their pockets most of their waking hours, it can act as a pedometer counting steps and detect when the user is sedentary. For runners who use running apps to track their runs, IMUs enable some apps to track indoor runs and cadence. Sports Engineering Researchers have used smartphones for tracking wheelchair rugby activities and classifying different sporting activities.

As great as the smartphones are with inbuilt IMU, GPS and processing power to give us real-time analysis, we don’t really want to strap an expensive smartphone onto a football player’s calf to monitor their kicking or tape an iPhone to a tennis racket to measure swing metrics. That’s why companies like Qlipp has developed sensors for tennis or Zepp which has sensors for a number of bat-and-ball or swing type sports. Then there are sensors for rowing, running, surfing, mountain biking and more. There are also different sports equipment that has in-built IMU sensors. Like smart balls (basketball, football, cricket ball etc), smart shoes, smart helmets, smart rackets etc, it could go on and on.

But sometimes we might still not find a sensor product on the market that is right for our sports or health application. So we explore the option of developing something on our own. Fortunately, we don’t necessarily have to start from scratch* because these days there are generic IMU sensor platforms that are designed and built for people who want to develop a sensor for a custom application. They often have the standard 9-DOF (degree of freedom) sensor setup and come with software SDK that allows developers to build their own applications for processing and analysing the data. Let’s look at a couple of options below.

[*when I say scratch, I mean getting sensor boards from SparkFun, Adafruit, Seeedstudio, Tindie etc]

mbientlab

mbientlab successfully launched their first Bluetooth IMU sensor on Kickstarter. They pitched it as a development and production platform for wearables with simple API for iOS and Android. There was some simple soldering required when people bought the first product. I didn’t get one from that campaign but I did get a later updated version which they called MetawearRG. What impressed me when I first got it was the size of it – it’s small and compact and I could use it to build/redesign a smart basketball prototype for a client. Then when I started testing it, I found that their API was really easy to use and I could use their sample iOS app to build a custom app for testing within a (reasonably) short time.

smart_ball_instagram_post
Smart Basketball Prototype and Watch app for tracking optimal shots

Since then, they have made many other versions of sensors with:

  • slightly different sensor configurations,
  • options of coin cell or rechargeable lithium battery,
  • accessories such as cases, clips or wristbands,
  • sensor fusion firmware,
  • cloud services, and
  • hubs to manage multiple sensors.
IMG_1624_edit
Metawear RG with custom 3d printed sleeve/case (L) and Metamotion (R)

I haven’t had the chance to try everything but I have to say, I have had a good experience using their Metawear and Metamotion sensors to build various proof of concepts and I am still using them for a number of projects. The sensor data can be streamed to your smartphone or logged on the device. In terms of API support, on top of iOS and Android, they have added Python, C, C# and Javascript, so developers can build stuff on various platforms.

IMG_1629

Sample/Template Metawear iOS app for testing

Looking at their new website revamp and some recent emails they sent out about new platform developments, they seem to be putting more focus into the allied health space, in particular, measuring range-of-motion (ROM). They are currently beta testing an app called the MetaClinic and it looks like they are using skeleton-tracking the likes of motion capture systems which would probably mean we need to use multiple sensors. That should be interesting.

mbientlab_metaclinic

MetaClinic App by mbientlab

Notch Sensors

Notch also launched on kickstarter, in fact slightly earlier than mbientlabs’ campaign. They had an interesting concept of integrating individual IMUs into custom designed clothing using pockets in discreet locations. Unfortunately, they weren’t successful at that instance. Their initial use case probably wasn’t strong enough. So I guess the founders went back to the drawing board, revamped it all and went with the “motion capture” approach for developers.

IMG_1627

Notch sensor with elastic band and clip

With the new design, the shape of the IMU sensor is essentially the same but they have ditched the micro-usb in each IMU for contact pins and made it water-resistant (IP67). They also designed elastic bands of varying lengths with a sensor clip and a user can secure each sensor up to 15 different locations on their body including head, chest, upper arms, wrists, hands, waist, thighs, ankles and feet. So instead of selling individual IMUs, they sell a kit of 6 IMUs with a set of elastic bands, and if a user wants to do a full (body) setup, they will need 3 kits.

IMG_1130

The Pioneer Kit: 6 IMUs with charging case and elastic bands with clips.

A quick test and review (for biomechanics)

I had the opportunity to run a short pilot test with one (the pioneer) kit in a biomechanics lab. I used the lower body setup which used all 6 IMUs strapped on my chest, waist, thighs and shins/ankles. In terms of setting up, it was pretty straightforward. After following an initial calibration procedure of all the IMUs in the case, I put on the bands and clipped each IMU to the right location according to the different colours as indicated on the app. The only thing is putting on the bands takes a bit of practice and I had to swing around to check that the bands are not too tight and restricting movement. Even though I don’t have muscly quads, I felt that the bands were somewhat tight and needed adjusting after a while.

WAOR1807

Setting up the Notch IMUs for lower body measurements

For testing, I did a simple protocol of walking, stopping and doing 3 squats of varying depths. Then I compared my knee angles measured on the notch and the motion capture system. A few quick things that I took out of the knee angle measurements were:

  • The angle measured by Notch is the exterior angle while the motion capture system looks at the interior angle. So it needs a quick recalculation before comparison.
  • Assuming the motion capture system is the more accurate measurement, Notch had a larger error as squats went deeper.
  • But for walking, the knee angles measured were quite close.

It’s wasn’t a very elaborate test but even from this simple outcome, I can safely say it’s probably not the best tool for accurate joint angle measurements. Although for a quick 3D visual feedback on movements, it might work. Here’s the clip of me doing the test described above (feel free to rotate the video to get different perspectives):



Further to that, I could only download angle data. If I wanted the raw sensor (acceleration and gyro) data, I would need to pay for an extended license that is renewed annually.

In terms of custom development support, they used to have support for iOS but they seem to have taken that off now and only have support for Android which I thought is a bummer. I am guessing they have some issues with getting it right on iOS. Hopefully, it is just temporal and they will resolve it soon. For Android developers, it looks like they have pretty good support and even provides a template app. I have to add that there is a fair bit of fine print I need to agree to before I can get access to their SDK. If I read it right, they basically want a licensing fee for using/commercialising their SDK.

Custom Sensors

Both of the above IMU sensors have similar specifications when it comes to measuring acceleration (using accelerometers) and angular velocity (using gyroscopes). The typical measurement range for accelerometers is +/-16g (that’s 16 times of gravitational acceleration), and for gyroscopes, it’s +/- 2000 degrees per sec. For many applications, this configuration is fine. But there might be some cases where higher acceleration needs to be measured and that goes beyond 16g, like shocks or high impact collisions. Or I might need high-speed rotations to be tracked and 2000 degrees per sec is too low, like measuring the spin of a cricket ball or gridiron football (which can come close to 3600 degrees per sec or 600rpm as demonstrated here by Drew Brees).

IMG_7657

Spin rates of a gridiron football during a throw test

As briefly mentioned earlier, hobby electronics stores like SparkFun, Adafruit, or Tindie would be a good place to start when looking for accelerometers and gyroscopes of different specifications. There are also lots of microcontrollers with Bluetooth Low Energy (BLE)  built-in that are Arduino compatible so we can program them with the Arduino software. One that I found pretty handy is this one called Blueduino which comes with a Lipo charger add-on (and add-ons are great) and that can be found on Tindie.

Football sensor

The gridiron football sensor prototype using the Blueduino

Final Word

For those who are in research and possibly need Matlab and software support for building custom Matlab programs, definitely check out Sabel Sense sensors (Australia). Else, I reckon the mbientlab sensors would be a great option for starting a custom development. If I get a chance to trial their Metaclinic platform, I will put up another post. Meanwhile, do drop me a message here if you need assistance or advice in any of the options above and feel free to leave a comment if you know of better/different solutions out there. With that, thanks for reading!

Do Force Platforms, Pressure Sensors And Smart Insoles Do The Same Thing?

Force platforms, pressure sensors and smart insoles are all devices that a person can step on and get some insight related to their weight or the pressure they are exerting on those devices with each step. Other than that, they are quite different and can have very different applications. This post is just an attempt to break that down. Feel free to jump to the different sections that are of interest:

[Force PlatformsPressure SensorsSmart InsolesSummaryMore on Smart Insoles]

Force platforms

A Force Platform (FP) is an equipment that you would typically find in a lab – an engineering lab, a biomechanics lab, gait analysis lab, ergonomics lab.. you get the idea. They are great for measuring forces applied directly onto its surface. So when a force platform is placed on the ground, you could step on it to find out how much force you are exerting on the platform. For those platforms that measure multiple axes, you could also slide an object across the platform to measure resistance forces between the surfaces. In sports engineering, FPs enable studies in walking/running gait, jumping (and landing), friction measurements in water polo balls or shoes or gloves, the coefficient of restitution of balls, aerodynamic drag (when placed in a wind tunnel), and more.

IMG_1075

An example of a Kistler Force Platform (blue) set up in a wind tunnel

For anyone keen to explore what else is done with force platforms in sports engineering, feel free to do a quick search on these journals: Sports Engineering JournalSports Technology Journal or Journal of Sports Engineering & Technology.

Inside Force Platforms

The majority of Force Platforms in the market are set up with multiple Strain gauges or Piezoelectric sensors/elements that deform proportionally to the applied load. There is also the not so common Hall Effect sensing Force Platform which doesn’t require an external signal amplifier/conditioner like the strain gauges and piezoelectric sensors do. They are typically quite expensive and their prices vary with the number of sensors, size, construction, and additional data acquisition (or signal amplifier) systems.

For those who can’t afford the expensive systems and is adventurous enough to try and build something, a sports physics researcher from the University of Sydney wrote a paper providing details of a cheaper home made force plate. Essentially he used Piezos that were manufactured for sonar applications and they cost $25 each. A quick search on Instructables also showed one DIY instruction on making a strain gage force plate. For the slightly less adventurous, there is also the option of the Wii Balance Board as a cheap force plate alternative. There have been some validations of the gaming platform as a standing balance assessment tool, a golf swing analysis tool, and for use in other medical applications. The only downsides of the Wii Balance Board are the user weight limitation and that a custom software is required to access and read the data.

Pressure sensors

There are three main differences between Pressure sensors and Force platforms. Pressure sensors are typically flexible and can be placed on flat or curved surfaces, unlike Force platforms that have to be mounted rigidly. The other difference is pressure sensors do not measure force vectors. Thirdly (or a slight extension of the second), Pressure sensors only quantify pressure that is perpendicular to it (single axis) so it cannot determine shear forces or friction between two surfaces. Due to their flexibility, pressure sensors have been used to determine comfort and fit in aircraft seats for Paralympians, analyse medical mattresses, measure the pressure of grip during a golf swing, pressure distribution on bicycle handlebars, and more.

fsr402

Single force sensitive resistor (FSR) from interlink electronics

Pressure sensors are mostly made out of either resistive sensors or capacitive sensors. The main differences between them are the sensing material used and their electrodes. They can be constructed as single sensing nodes or they can also be constructed in a row-column array fashion. The advantage of the array or matrix construction (over single nodes) is that it requires fewer connections. In an array, the intersection between each row and column is a sensing node. So a 3 by 3 array creates 9 sensing nodes while only needing 6 connections.  On the other hand, 9 single sensing nodes will need 9+1 connections where the +1 is the common ground. The difference becomes much bigger as the number of sensing nodes increases (For example 100 sensing nodes can be achieved using a 10 by 10 array that needs 20 connections or 100 single sensing nodes that need 101 connections).

Single Sensor Nodes Vs Arrays 2

A simple illustration of Single sensing node Vs Sensor Matrix/Array

However, the matrix construction is not without its challenges. The matrix sensor circuit is prone to parasitic crosstalk (capacitive or resistive). This means when pressure is applied on one node or multiple nodes, the electrical readings for other (unactivated) nodes might be affected. This is also known as “ghosting”. Unless some correction is applied, the measurements/readings become inaccurate and potentially useless. Also, the bigger the matrix, the more complex the correction. But if accurate absolute readings are not required, then it’s fine.

A related side story

I have been following the development of this smart yoga mat that was successfully crowdfunded on Indiegogo back in Dec 2014. Fast forward to 2017, they are still struggling to deliver the product. Looking through their updates, we can see they had to deal with sensor accuracy (possibly the crosstalk or ghosting issue); and on top that, some other issues they had include sensor durability, mat materials suitability, and accuracy of their tracking algorithms (which they are using some form of AI). Having prototyped a smart exercise mat around the same time they started, I can fully understand the challenges and why it is taking that long. Then again I am not sure it is worth all that effort. Personally, I think that simply relying on a pressure sensing mat to monitor and give (technique) feedback on yoga poses (or any exercises) has its limitations. Adding camera tracking (possibly utilising the camera on the tablet) might help. That saying, it is not stopping others from developing similar products as seen in this video.

Smart Insoles

Smart Insoles or Instrumented Insoles are essentially pressure sensors made in the shape of a shoe sole. The sensors are usually made in a similar fashion described earlier. Most of the Smart Insoles are also built with IMUs so that it adds a bit more context to the pressure data such as whether the wearers are standing, walking, running or jumping. The greatest advantage of Smart Insoles is they allow feet pressure mapping and measurement on-the-go. Things like continual gait analysis and activity monitoring, and it even has medical application likes foot ulcer prevention and falls prevention.

contnt_img

Source: Footlogger.com

There are a couple of shoemakers that designed their shoes with the Smart Insole embedded within the shoe like the Altra IQ for running and the Iofit for tracking golf swing stance. The good thing about them is they have designed everything to fit properly into a shoe, made for a specific function. So users don’t run the risk of their Smart Insole not fitting properly into their shoes and collecting inaccurate measurements. On the other hand, users are restricted with specific shoes for pressure monitoring or activity analysis.  But at the end of the day, the pros and cons are really dependent on the individual.

Brief Summary

Going back to the question: “Do Force Platforms, Pressure Sensors and Smart Insoles do the same thing?”; there are some things that they are all capable of performing (e.g. gait analysis), but they all do it in a different way.  Also, there are certain measurements or monitoring that are unique for each sensor. Here’s a simple table that sums it up:

Sensors Measures shear force Measures Pressure Doesn’t require rigid mounting Portable Tracks Motion
Force Plates X X ✔/X
Pressure Sensors X ✔/X X
Smart Insoles X


More about Smart Insoles

Personally, I feel that Smart Insoles is a great idea, with many useful applications in sports and health. Over the last few years, there has been an increase in research and development in this area with many patents generated in the process; and companies around the world have come up with commercial products around the concept of Smart Insoles. It is definitely still in its early stages and I am not sure if it has even reached Early Adopters yet. Sadly, one company that I followed (Kinematix) has already closed shop due to a lack of funding. Perhaps it is ahead of its time like the adidas intelligent running shoe with intelligent active cushioning. Nevertheless, I believe the potential (of Smart Insoles) is there and I think targeting specific niches/problems will probably have a better outcome than designing for a generic application.

If you have an idea or project needing a smart insole or custom pressure sensor, feel free to contact us or leave a comment. We might be able to help you with it or at least point you in the right direction. As always, thanks for reading!

This post also appears on sportstechnologyblog.com: link.


Other related articles:

Tracking & Managing Anxiety in Athletes Using Wearables

The 2016 Rio Olympic games as with the previous games was a great platform for many tech companies to showcase their latest developments. There are radar and camera technologies that capture motion/biomechanics of an athlete on the field and in the pool. There are wearable devices that (also) track motion plus monitor physiological parameters 24/7. They aim to positively alter athlete behaviour and optimise performance. There are also sports apparel and equipment that were designed and developed (after much R&D) to enhance athlete performance. But we will leave that for another time.

Wearables for tracking performance

Going back to wearables and tracking systems; they often look at (somewhat) straightforward parameters – joint positions, speed (or velocity), height, acceleration, impact, angles, rotation rate, heart rate, heart rate variability, sleep and other physiological stuff. Sometimes coaches and athletes only need to look at a single parameter while other times they may need to examine a combination of variables and find correlations or visualise them over time to identify trends. Some companies go further by processing the above data and coming up with (trademarked) indexes such as Player-Load (Catapult), Windows of Trainability (Omegawave) and Recovery Score (Whoop). What they are trying to achieve is break down all the data that is being collected and deliver one metric that simplifies things and make it easy for coaches and athletes to measure performance (and recovery) .

In major games like the Olympics, where athletes trained years to prepare and qualify for that one event and possibly one moment, there can be a lot of anxiety and pressure to perform. Even if all the physical preparation has been done right, the results could still boil down to how well those emotions are managed; the difference could be between a podium finish or not performing as well as expected. So are there wearable technologies that monitor an athlete’s emotions and maybe warn the athlete of dangerous anxiety levels that can lead to choking or panic?

Wearables for tracking anxiety

Turns out there are a number of wearables in the market that do that. Here are three different types:

  1. Head-worn wearables that measure EEG signals (or brain activity) like the Emotive Insight and Muse. Although the Muse is designed as an aid for meditation and relaxation, it is basically monitoring four EEG channels to see how excited or relaxed a person’s brain is. The Emotive Insight has five EEG channels and looks at the user’s cognitive performance in areas such as Engagement, Focus, Interest, Relaxation, Stress, and Excitement. Emotive also has a higher spec neuroheadset that can look at fourteen EEG channels and goes into much more depth of what’s going on in a person’s mind and how he/she is feeling.

    emotiv_epoc_600

    Emotiv Epoc+: 14 channel wireless EEG system

  2. Wrist-worn devices that measure electrodermal activity (or EDA), blood volume pulse, skin temperature and motion; like the Feel and Empatica E4 wristbands. Based on research, measurements of EDA strongly reflect sympathetic activation which is linked to stress levels and excitement. Measuring heart rate variability through the blood volume pulse sensor also reflects sympathetic and parasympathetic activation. Skin temperature is another reliable measure of stress levels as shown in this research. Finally, motion tracking with inertial measurement units (or IMUs) helps identify the user’s activity and tries to place a connection between anxiety levels and what the user might be doing at that time.

     

     

    Screen Shot 2016-08-30 at 10.19.16 AM

    The Empatica E4 and Feel: 4 sensors packed on a wrist device

     

     

  3. Clipped-on devices that measure breathing frequency like the Spire. The Spire is built with force sensors; when it is secured onto the user’s waistband or bra, it detects the expansion/contraction of the user’s torso and diaphragm during breathing, thus deriving the breathing rate. Then algorithms are used to determine from the breathing waveforms whether the user is calm, tensed or focused.

 

Screen Shot 2016-08-30 at 12.30.42 PM

Spire: Breathing frequency tracker

 

Most of these devices also provide an accompanying app to monitor anxiety levels, and they prompt users to meditate or do breathing exercises. On a side note, a breathing exercise for lung patients was adapted for training athletes’ breathing technique and also focuses on dealing with anxiety. Athletes could also listen to brain.fm music that either helps them relax or stay focused. In a way, managing stress levels on a day-to-day basis can be beneficial for athletes because stress levels can increase the likelihood of an athlete falling sick or getting injured, and it also affects recovery.

Emotion Profiling for Performance

On the other hand, when it comes to performing well during competitions/races, some athletes actually perform better with some amount of anxiety. In fact, different athletes in different sports may perform better at varying levels of anxiety. In other words, some athletes perform well at high levels of arousal while others may perform better at lower levels of anxiety. It’s all about finding a sweet spot. As mentioned in this article, one widely used tool by coaches/athletes to identify that sweet spot or optimal performance zone is the individual zones of optimal functioning (IZOF) model. This is a qualitative analysis approach that involves the athlete recounting the emotional experiences related to successful and/or poor performances. All the emotions are then labelled and rated as described here, and this creates an individualised emotion profile showing which emotions are helpful for performance and which ones are unhelpful. Of course, this would only work if athletes have competed for a number of times previously and came out with different outcomes (winning or losing or setting new personal bests).

Performance-arousal-IZOF_2

Individualised emotion profiling (source: sportlyzer)

Ultimately we could utilise all the different wearables (and tools) mentioned above and somehow piece all that data together to shed some light on the inner workings of each individual athlete. Then the data could be used to “pivot” them in the optimal direction. But at the end of the day, its really down to the athletes themselves pushing hard every day and fighting battles with their body, mind and soul to get to where they would be. So let’s just salute the Olympic athletes for what they do and what they have achieved. And while we await the start of the Paralympics, I leave you with this video below by Under Armour and Michael Phelps. Thanks for reading!

Some thoughts and takeaways from #SAC16

SAC16Banner

The 2016 Asia-Pacific Sports Analytics Conference took place recently at the NAB Village. Its only the second time this conference is held and I have to say it has done really well. The numbers prove it – 865 attendees (according to the Whova app), 33 sessions that ran concurrently in 3 different rooms, 45 Speakers (all experts in their fields) representing 57 organisations, and 12 startups that pitched their innovative ideas/products/services.  There was even a waitlist 2 weeks before the event. This goes to show the growing booming popularity of data analytics, and the potential impact it could have on the different aspects of sports.

TNRE2583

You know it’s a serious conference when it has its own coffee cup

Unfortunately, as with any great conference where there are sessions running at the same time, people would be torn between 2 (or possibly 3) presentations they are keen to attend. Fortunately, from what I heard, videos of all the sessions will be uploaded in a few weeks and we will be able to catch up with every single one that we missed. Just keep a lookout on the conference website here. In the meantime, here are some of my takeaways from the few sessions I managed to attend.

Smart equipment:

Professor Tino Fuss presented some of the research and development that was going on at RMIT including a smart cricket ball, a smart soccer boot and smart compression garment. With the advancement of inertia sensing microtechnology and novel pressure sensing technology, sensors can be placed unobtrusively on the athlete and equipment, measuring a range of parameters at much higher magnitudes. No doubt that the sensor data that’s acquired has to be analysed to solve a problem or confirm a hypothesis. That’s where analytics play an important role. But applying the appropriate sensor technology does open up opportunities to analyse new parameters like the sweet-spot on a soccer boot that increases the chance of a goal.

Wearable tech for rehab:

Shireen Mansoori is a doctor in physical therapy who applies wearable technology in her practice with elite athletes. She presented a model where she combined physiotherapy and data analytics for athlete optimisation. She uses Catapult units for monitoring an athlete’s Player Load & Hi Deceleration efforts to find trends that lead to injury. But she also uses other wearable tracking devices such as the Misfit shine on the athletes, health/wellness monitoring apps, and an athlete sleep screening questionnaire to monitor an athlete’s sleep and daily activities. Having other forms of data paints a much clearer picture of what an athlete is going through, and allows her to find out why the athlete is recovering faster or performing below expectations.

Video analysis & Artificial intelligence:

In cases where it is still obtrusive to place sensors on athletes (for example in swimming competitions); or where wearable sensors can’t provide specific activity/events information (for example attack, pass or steal events in hockey), sports analysts turn to video analysis/coding. However, much of the video analysis work involves a sports scientist (or two) manually tagging/coding every event during the competition. Stuart Morgan, sports analyst at AIS, talked about developing computer vision algorithms to  detect patterns and features and somehow automate the tagging. But this approach (human engineered method) has lots of limitations including it being non-transferrable and not very adaptable (for use in different sports). So AIS is collaborating with researchers at La Trobe Uni to apply deep learning (using Convolutional Neural Networks) to process the video images and work out whats happening. The advantage of deep learning is that it’s adaptable and it automatically creates new features. It still has some way to go as it’s not error free and users can’t really tell what logic led to the decisions.

IMG_3831

Stuart Morgan talking about AI in sports analytics

From elite to grassroots:

Most of the stuff mentioned above happens in the professional/elite athlete space. However there is also an increased trend of sports tech/analytics companies developing products for athletes and coaches who participate in their local leagues. Hudl‘s video analysis software was first developed for professional teams. But today, their software caters to high school teams and their requirements. They have developed mobile apps that allows video recording and editing directly from the coaches’ mobile device, and there’s even a platform for sharing videos and facilitating talent identification.

Athlete tracking wearables have also moved in the same direction. Startup companies like Essential GPS and Sports Performance Tracking have developed more affordable tracking solutions so that teams with lower budgets can also track and monitor their players. Although it seems to be purely GPS data (without motion data), and only post game/training analysis (not real-time), it is still a good start. Or maybe a simplified, cost reduced system is all that is required?

From the startup community:

So there were 12 startups showcased in the conference. Other than the 2 mentioned above, there were 4 other startups that have built hardware in areas of performance tracking, drone racing, rehabilitation, and custom protective gear. The others were mainly software based, providing services and platforms in media, news, sales, marketing, VR and team management. They have all developed solutions hoping to fill a gap identified in the sports industry. Personally I am just amazed at some of the novelty and innovation they have come up with; and as this blog post says it, they are all innovators.

Bottom line:

I think what sums up this conference for me is that sports analytics is all about adapting and innovating. Everyone in their own ways are trying to fix a problem (or come up with a better solution) or improve work flow or even create new opportunities (e.g. esports and fantasy league). But the process is never a straight line from point A to B. The solutions need to be adapting over and over (almost like deep learning). Sometimes there needs to be collaborations and sometimes the end solution needs to be a combination of solutions. Whichever the case, iterate the process as quickly as possible till an optimum outcome is reached.

The”one-size-fits-all” solution doesn’t work very well anymore and mass customisation is becoming the norm. As mentioned by John Eren MP and Laura Anderson during their welcome addresses, we are slowly moving away from economies of scale and towards economies of scope.

SAC16_welcomeAddress

Group photo after welcome address. From John Eren Mp’s facebook page (link)

Anyway, congrats again to PSCL and KPMG for another successful event and thanks for reading!

Of Racing Suits and Aerodynamics

Wind Tunnel tests with custom designed mannequins and different Under Armour speed skating suit prototypes.

In many sports that involve high speed movements, drag or air resistance is probably one of their biggest enemy in achieving their peak performance. One winter sport that faces this challenge is speed skating, and turns out altitude plays a big part as well – the higher the skating venue is, the less air resistance there is (more about that in this article). Also the effect of drag on the skater’s speed and performance is pretty significant and the suit that the skaters wear could have an impact on the colour of the medal they get.

So just before the 2014 Sochi Winter Olympics, there was a bit of news about the revolutionary speed skating suit designed and made by Under Armour and Lockheed Martin. The “Mach 39” was supposed to be the fastest speed skating suit ever made. Unfortunately, instead of delivering medals (gold ones for that matter), the result was the US athletes performed below expectations. Now, this could be due to the suit OR if we break it down, could be due to a thousand other reasons (on top of the suit)..

There was a bit of history to the design of the suit, and the basic idea was: just as dimples on golf balls reduced aerodynamic drag, adding dimples on the suit would have the same effect. Of course, other than the dimple design, there were other considerations like textile selection and compression fitting design. Just have a look at the video below that describes what the designers and researchers looked at to reduce friction and improve aerodynamics of the suit. What’s really interesting is how they customised the mannequins to typical skating positions for wind tunnel tests. (Drag to 4:00 of the video to just see the custom mannequins)

Although the rational behind the design and testing all seems to make sense, I can’t help but have a few questions:

a. With so much movements during speed skating, is it really possible to estimate the drag based on wind tunnel experiments? I mean, there are a number of sports that do drag tests in wind tunnels; like skiing and cycling. But these sports have moments of competing when the athlete maintains a certain position for a short period; and those are the moments where having an optimum position (aerodynamically) could really reduce drag significantly. But speed skaters hardly stay in one position during competition (maybe except at the starting line). Then if that’s the case, would the wind tunnel results be fully applicable on the track?

b. Friction plays 2 roles: it slows you down and it gives you more grip/control. If there is too much friction, it impedes movement; but if there is minimal or close to no friction, the athlete might lose control. How then, do we strike a balance between them?

c. Is it possible to measure drag dynamically on the track? Well, a company called Alphamantis seems to have done that, but with cycling, and in a velodrome fitted with gate sensors. Some additional input parameters they require include the bike’s wheel circumference and also inputs from standard power meters and speed/cadence sensors. With the power meters, there is a calibration process before the actual aerotesting where they apply a model to calculate drag. For more details of the testing, you can read ths interesting blogpost by DCrainmaker.

I reckon it is possible (in theory) to develop a model for speedskating (similar to what Alphamantis did for cycling) to estimate drag on the ice skating track. The model might be slightly similar to this one in wheelchair racing: when the speedskater is pushing off (and at equilibrium), there are 4 different forces applied on the speedskater: 1) Reaction force, 2) Inertia, 3) Friction between the ice and skates, and 4) Drag force.

  1. Reaction force (or applied force) can be measured by instrumenting the skates with a shoe sole pressure sensor similar to this or this.
  2. Inertia can be determined by measuring the forward acceleration of the skater (using an inertia sensor or a suit of sensors), then multiplying that by the overall mass of the skater.
  3. Friction can be calculate based on the coefficient of friction of ice which is different for straights and curves according to this paper.
  4. Finally, since the sum of all these forces equals to zero, we can determine the drag force!

Xsens Concept Tests in Speedskating

Of course this model is very much simplified and some assumptions are made, but if more thought is put into it, this might just work.

Anyway, going back to the lacklustre results of the Under Armour Mach 39 suit, there could be so many reasons why the athletes didn’t perform during those races. Since US speedskating has extended the contract with UA, they obviously know that the suit wasn’t the main culprit. It did sound like the athletes weren’t really used to the new suit, so maybe it’s just a matter of ‘breaking-in’ the suits.

Thanks for reading and if you have any thoughts or suggestions on aerodynamics or drag tests, do leave some comments!

(Also posted in SportsTechnologyBlog.com)

Wheelchair Rugby: First Paralympic Gold and some Breakthrough Research

Photo from the London 2012 website

The Wheelchair Steelers (Australian Wheelchair Rugby team) recently won the much coveted gold medal at the London 2012 Paralympics. The Steelers have been aiming for this gold medal since the 2008 Beijing Paralympic games where they got silver. Over the past 4 years, the coach and team have been working hard to hone their skills, establish training centres and at the same time attract new talent. There has been year round trainings and competitions, international ones such as the 2010 WWRC, and locally, there’s the annual National League, and annual State League in Victoria. Organisations such as the Disability Sport & Recreation have also been promoting the sport through programs and social media to encourage more people to try out the game.

On the research side of things, there has been a couple of studies conducted at the RMIT Sportzedge program with support from some of the Australian wheelchair rugby athletes and coach:

1) Customisation of rugby wheelchairs for performance – The idea of performance based customisation is to maximise the athletes’ comfort and performance through adjusting a few key parameters of the wheelchair design and finding the optimum setup. The main experiments were designed after much research and field tests were done and this was of course coupled with feedback from the athletes and coach. In the end, a platform was developed that allows athletes of the various classifications to systematically customise their individual wheelchair that not only feels good but also helps the athletes perform. For a video on the wheelchair customisation research, check out Channel Ten’s program Scope where it was featured in an episode on Science in Sports.

Ergometer Tests with the Wheelchair Rig

2) Performance & match analysis of wheelchair rugby athletes using inertial sensors – Match analysis of wheelchair rugby was motivated by the fact that the only option available currently is video software analysis, and even though inertial sensors are so commonly used in other able-bodied team sports, it hasn’t been applied in wheelchair sports. The challenge however is to use the kinematic data for activity identification, and not just for measuring speeds and accelerations. The final outcome would be to use the likes of smart phones that are embedded with MEMS sensors and are programmable, mount them on the rugby wheelchairs during competition, and run apps that can determine and track the various activities and performance. Ultimately, this could assist the coach in monitoring the athletes’ performance or even be used for disability classification studies.

Below’s a list of publications that resulted from related work done in the past 3 years. Most of them were presented at the previous ISEA and APCST conferences as well:

Although most of these work were focused on wheelchair rugby, the concepts and platforms developed could potentially be applied to other wheelchair sports for user optimised wheelchair designs and for monitoring activity & performance.