Automated Basketball Shooting Trainer Using Computer Vision
Jerzees Fernandez, Abdulaziz Abdullah, E.J. Diao, Trenton Tong-Seely
School of Engineering
Faculty Supervisor: Hamid Mahmoodi
This project focuses on the design and development of a low-cost automated basketball shooting trainer that assists players during shooting practice. Commercial basketball training machines are often expensive and provide limited customization, making them inaccessible for many students and amateur athletes. The goal of this project is to develop an affordable system that can track a player’s position and automatically pass the basketball during training drills.
The system integrates a smartphone equipped with a camera and LiDAR sensor to track the player in real time and measure the distance from the basket. A Raspberry Pi processes the tracking data and calculates the required passing force. The launcher mechanism uses high-power motors and a servo system to rotate and deliver the ball toward the player from different angles. Gesture recognition is also incorporated to allow the player to control commands such as pass, pause, or stop.
By combining computer vision, distance sensing, and motor control, this project demonstrates a practical and cost-effective approach to creating an intelligent basketball training system. The prototype provides a foundation for future improvements in automated sports training technology.