Symposium

Of Rising ScholarsFall 2022

Akshay will be presenting at The Symposium of Rising Scholars on Saturday, September 24th! To attend the event and see Akshay's presentation,

Register here!
Go to Polygence Scholars page
Akshay Shivkumar's cover illustration
Polygence Scholar2022
Akshay Shivkumar's profile

Akshay Shivkumar

Mountain View High SchoolClass of 2024Los Altos, California

About

Projects

  • "Considering the options of using machine learning and electrical engineering with an Arduino and sensors, what is the most efficacious method to predict a tennis court's occupancy?" with mentor Beck (Working project)

Project Portfolio

Considering the options of using machine learning and electrical engineering with an Arduino and sensors, what is the most efficacious method to predict a tennis court's occupancy?

Started Apr. 4, 2022

Abstract or project description

Whenever people go out to play tennis, they often find themselves presented with a frustrating problem; all of the courts are occupied. Thus they must venture out to the next court near them to see if a court is unoccupied there. Even big parks with many tennis courts would be busy with players, lining up for their turn. None of the courts have a way to check availability or reserve it. Hence, we investigated various options to find the most efficacious and reliable method to predict a court’s occupancy and inform people. We found that using machine learning models to classify noise as hitting a ball would prove as one effective way to predict if someone is playing. We also found that using motion sensors along with an Arduino proved fruitful, as this can sense if there is a player on the court. The third option is to use a Bluetooth proximity sensor, which tells us if there is a phone or device on the court, signifying that there is a player as well. To test which one is the most effective, we will set up each of the three systems and measure how accurate and reliable they are at predicting occupancy through a series of tests. Our results will tell us which method is the best and will allow us to build a full prototype of the system. This could have broader implications, as we may be able to implement it in the tennis courts and build a website for everyone to use.