Optical and Acoustic Sensor-Based 3D Ball Motion Estimation for Ball Sport Simulators †.

Area of Science:

• Robotics and Automation
• Sensor Technology
• Sports Science

Background:

• Accurate estimation of 3D ball motion is crucial for realistic ball sport simulators.
• Existing camera-based methods have limitations in terms of activity area.

Purpose of the Study:

• To develop and evaluate a novel system for estimating 3D ball motion, including speed and projection angle.
• To overcome the limitations of camera-based systems by utilizing acoustic and infrared sensors.

Main Methods:

• A three-step system: sound-based ball firing detection, sound source localization, and infrared (IR) scanning for motion analysis.
• Impulsive sound classification using mel-frequency cepstrum and feed-forward neural networks for launch detection.
• 2D microelectromechanical system (MEMS) microphones and delay-and-sum beamforming for sound source localization.
• High-speed infrared scanning for determining the ball's 3D position and trajectory.

Main Results:

• The proposed system successfully detects ball launch sounds and localizes the firing position.
• The system accurately determines the time and 3D position of the ball.
• Experimental results show a wider activity area compared to camera-based estimation methods.

Conclusions:

• The developed acoustic and IR sensor-based system provides effective 3D ball motion estimation.
• This technology offers practical applications for various sports simulations, including soccer and baseball.
• The system's expanded activity area enhances the realism and usability of sports simulators.