Reliability of a Low-Cost Inertial Measurement Unit (IMU) to Measure Punch and Kick Velocity

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 2; p. 307
• Main Authors: Pezenka, Lukas, Wirth, Klaus
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.01.2025; MDPI
• Subjects: Acceleration; Accelerometers; accelerometry; Accelerometry - instrumentation; Accelerometry - methods; Adult; Algorithms; Analysis; Athletic Performance - physiology; Boxing; Business performance management; Cellular telephones; Combat sports; Economic aspects; Humans; Kickboxing; kinematics; Male; Martial arts; Martial Arts - physiology; Measurement; Motion capture; reliability; Reproducibility of Results; Sensors; Smartphones; Software; Velocity; Warm up (exercise); accelerometry; kinematics; combat sports; reliability
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s25020307

Striking velocity is a key performance indicator in striking-based combat sports, such as boxing, Karate, and Taekwondo. This study aims to develop a low-cost, accelerometer-based system to measure kick and punch velocities in combat athletes. Utilizing a low-cost mobile phone in conjunction with the PhyPhox app, acceleration data was collected and analyzed using a custom algorithm. This involved strike segmentation and numerical integration to determine velocity. The system demonstrated moderate reliability (intraclass correlation coefficient (ICC) 3,1 = 0.746 to 0.786, standard error of measurement (SEM) = 0.488 to 0.921 m/s), comparable to commercially available systems. Biological and technical variations, as well as test standardization issues, were acknowledged as factors influencing reliability. Despite a relatively low sampling frequency, the hardware and software showed potential for reliable measurement. The study highlights the importance of considering within-subject variability, hardware limitations, and the impact of noise in software algorithms. Average strike velocities exhibited higher reliability than peak velocities, making them a practical choice for performance tracking, although they may underestimate true peak performance. Future research should validate the system against gold-standard methods and determine the optimal sampling frequency to enhance measurement accuracy.