IIT Bombay’s new web application, IMPART, allows researchers to track changing water surface temperatures and can help to track climate change

Playing the shots right

Read time: 1 min
Mumbai
2 May 2018
Photo : Ashutosh Raina

IIT Bombay researchers design a badminton training system using wearable technology.

Watching the world number one, K.Srikanth, play badminton is awe-inspiring! If you have ever wished to hold forth a rally like him, but had trouble practising the basic shots, then this study from the Interdisciplinary Program in Educational Technology at the Indian Institute of Technology Bombay could be good news for you. The researchers have designed a training system to help you master the shots in badminton using a Fitbit-like wearable device to record your arm movements and provide feedback.

In a fast-paced sport like badminton, one can play the right shot at the right time only if one knows it exceptionally well. To practise a shot to perfection, along with the stance one needs to learn when to apply the effort and how much to swing the arm. Perhaps, we can carefully watch experts play and imitate them to acquire these skills? That is not so easy! Effort and arm swing are abstract concepts which are difficult to explain and understand.

But what could help is a way to know the exact difference between one’s current shot and the expected shot as played by the experts. Here's where the training system, called CoMBaT, developed at IIT Bombay, suits well. Along with your coach, the system can help you practise at your own pace and schedule, and teach you how to improve each shot.

CoMBaT uses a band worn on the forearm, called the Myo band, developed by Thalmic Labs. It contains sensors that record the linear and angular motion of the player's arm and the muscular activity involved for each shot. The device transmits the recorded data wirelessly to the training system over Bluetooth, which is then used to calculate the swing of the arm and the effort applied for playing the shot. The training system also plots graphs for these parameters in real time which could be projected to an external display. Each plot contains a reference pattern obtained from the shots played by an expert, thus letting the player compare their shot with that of an expert’s.

Since it is not always convenient for a player to look into an external display while playing, the Myo band also contains vibration motors that provide real-time feedback by generating specific patterns of vibrations based on the quality of the shot. For example, if the player’s score for both swing and effort, calculated using the data received, are below a particular threshold score, the shot is ‘good’. The Myo band sends a short pulse of vibration, and visual markers  for swing as well as effort are in green on the external display. However, if one of the scores is above the threshold, the shot is ‘average’, and three short pulses of vibrations are felt. The corresponding visual marker  is shown in red. If both scores are above the threshold, the shot is ‘bad’, and the Myo band sends a long pulse of vibration with both the visual markers  shown in red.

While designing, the researchers studied the system with three novice players and one expert where each player played 50 shots. “We observed that the feedback on a shot, received from the Myo band, improved the player’s technique in the next shot”, say the authors of the study. “The visual plots gave immediate feedback on whether a shot was correct or not”, he adds, talking about the working of the system.

The study also gave some insights on the hidden aspects of training for novice players. For instance, the visual plots showed that muscular effort is applied before swinging the arm when playing a correct shot. While this aspect is evident to expert players, visualising it on a screen reassures the novice players about their technique.

The researchers plan to extend the functioning of the training system by better relating the plots to the actions performed. “The system could detect and guide corrections based on the common error patterns", say the authors, pointing out a few things they hope to improve. "Implementing virtual or stick figure animation for visualisation is also in the pipeline”, they say.

This research is a part of the activities of Next Education Research Lab at IIT Bombay, where researchers explore the uses of emerging technologies in the design of learning activities. The study is a contribution to the area on using wearable devices to provide meaningful avenues for teaching and learning. “Similar systems can be used for any sport that deals with the movement of arms and muscular force, such as cricket and golf, or activities such as pottery and carpentry. Further studies could explore training systems for leg muscles or foot movements such as in football and athletics”, sign off the authors, talking about the applications of such systems.