A team of IRL researchers are exploring the possibility of bringing IoT (Internet of Things) to sports analytics, particularly to the game of Cricket. Ph.D. candidates Mahanth Gowda, Sheng Shen, Ashutosh Dhekne and Prof. Romit Roy Choudhury developed solutions to track a ball’s 3D trajectory and spin with inexpensive sensors and radios embedded in the ball.
However, unique challenges arise rendering existing localization and motion tracking solutions inadequate for applications in fast-moving games, for example to track the ball in a Cricket match.
The system, termed iBall, developed at UIUC mitigates those problems by fusing disparate sources of partial information – wireless, inertial sensing, and motion models – into a non-linear error minimization framework, an advanced robotics technology that now might find its place in the sports industry. Compared against the millimeter-accuracy ground truth measurements provided by motion capture instrument hosted at IRL, the median ball location error can achieve a centimeter accuracy while rotational error remains below 12 degrees even at the end of the ball flight trajectory. The quality of tracking delivered by iBall is well-suited for applications in sports. This work has been accepted for publication on NSDI 2017, a highly competitive conference.