Estimation of baseball bat trajectory during a practice swing using a Kalman filter for velocity compensation

Abstract

This study aimed to (a) develop an algorithm that could estimate a baseball bat trajectory from the beginning of the swing to the follow-through phase during a practice swing without a ball and (b) evaluate the accuracy of the proposed method using a three-dimensional motion capture system. The sensor fusion using the adaptive Kalman filter for compensating velocity decreased the error of acceleration integration during the follow-through phase. Further, the three-dimensional bat trajectory in a global coordinate was estimated by combining the sensor fusion and compensation by motion characteristics. The three-dimensional bat trajectory from the swing beginning to the follow-through phase estimated by the proposed method was compared with the three-dimensional bat trajectory obtained by the three-dimensional motion capture system. The proposed method achieved a root mean square of the error of 7.72 km/h for velocity, which was less than the root mean square of the error (8.91 km/h) obtained by simple time integration of forward direction. These results indicate that the error by acceleration integration during the follow-through phase is compensated. The proposed method is, thus, deemed effective and can be used to evaluate baseball swing, including the follow-through phase, with high accuracy.

A Semantic Web-Based Approach for Bat Trajectory Reconstruction With Human Keypoint Information

“…During the authors' investigation, they did not see other bat trajectory reconstruction methods under similar conditions and scenarios. There are some studies on baseball (Shibata et al, 2023) and tennis bat trajectory estimation (Furuya et al, 2021) or some table tennis robot (Ji et al, 2021). There are also some studies on the ball trajectory reconstruction like badminton ball (Liu & Wang, 2022) and table tennis (H. .…”