For the final capstone project for my engineering class, my partners and I made a shot put distance tracker that could be used at high school track and field meets. Our targeted audience were high schools, because we wanted there to be an accurate, cost efficient, and accesible way to measure the distance the shot put travelled without human error.
My teammates and I decided to work on a solution for a problem that one of my good friends faced at high school track and field meet: inaccurate distance readings for shot put. She was a varsity shot put athlete, and she noted that a referee would estimate where the shot put would land, which wasn't always accurate as there wasn't always sand at meets or the ball would roll after it landed.
To determine if other shot putters felt the same problem, we surveyed 101 shot putters who competed at the high school and the college level. Over 40% of our responders have competed in the sport for over 5 years, showing we talked to an expereinced group. From this group, over 96% said that the distance is measured with an eye estimate and a tape measure. We asked them that on a scale from 1 to 10 how accurate and how consistent did they feel the measurements were, and we received an average response of 3.7 and 3.22, respectively.
We came up with a myriad of ideas: live motion profiling, use of photo and motion sensors, sound waves, a spray on solution. We attempted to come up with any idea we could, no matter how strange it seemed.
From our decision matrix criteria, we chose to pursue the motion profiling solution. We did a distance calculation for how our camera would need to interpret the shot put distance, and we created a flow chart of how we needed to code our Jevois cmsart-vision camera. We pciked this camera because we believed it was the most accessible and cost-effective for our target audience.
In this phase we mainly tested the camera on the ground to see how to best track the shot put. We coded different views of the shot put and tested it to determine what the camera would read best. We determined what the best Hue, Saturation, and Value (HSV) settings were.
We noticed that our results were skewed because out camera only rested on the ground, so we were getting some wildly inaccurate results.
My teammates and I then made a wooden block to rest on top of the shot put ring to enclose the camera there, and so it could have a better visual of shot put movement in the field. ALl that would have to be done is add the block to the ring, add the camera, and finally cover it. The distance would then be shown on the display window.
Overall, our results would typically be within 2 inches of the actual value that was measured using human eyesight and a tape measure. While we were still off in our set criteria for accuracy, we determined that with a better camera, the readings would be much more accurate. This solution would provide schools the opportunity of an accurate and cheap alternative rather than using human eyesight or expensive laser measurement equipment.
return to projects page