Validation of a LiDAR-based player tracking system during football-specific tasks

Abstract

AbstractTracking players’ movements in sports is important to monitor and optimise exercise volume, avoid injuries, and enhance game performance. A new LiDAR-based system (Sportlight®) purports to provide accurate velocity and acceleration metrics derived from player movements. This study examined the validity of the LiDAR-based system against a 3D motion analysis system. Two competitive football players (age: 18 years, height: 1.74 ± 0.01 m, mass: 66.5 ± 7.8 kg; playing experience at this level: 3 years) completed nine trials each of six sport-specific movements, consisting of straight-line sprints, cuts, and curved runs. Trials were recorded concurrently by a four-unit LiDAR system and a 64-camera 3D motion analysis system. Instantaneous velocity and acceleration, and time spent within key performance indicator bands (defined by velocity and acceleration thresholds) were compared between systems. Agreement between the systems was evaluated by root mean square error. Differences in time spent within each key performance indicator band between systems were assessed with t tests and standardised effect sizes. Velocity root mean square error values ranged from 0.04 to 0.14 m·s−1 across all movements and acceleration root mean square error values ranged from 0.16 to 0.7 m·s−2. Differences between systems for time spent within each key performance indicator band were mostly trivial. These results show that the LiDAR-based system can provide valid measures of velocity and acceleration in football-specific tasks, thus providing accurate tracking of players and calculation of relevant key performance indicators.