Mendeley  

Zotero  

Haj Lotfalian M. Validation of a Low-Cost Inertial Sensor for Estimating Barbell Velocity During Deadlift Concentric Phase. J Sport Biomech 2026; 12 (1) :20-34
URL: http://biomechanics.iauh.ac.ir/article-1-418-en.html

Objective In velocity-based resistance training, accurate measurement of barbell velocity is crucial for load prescription, fatigue monitoring, and autoregulation. Given the limitations of reference systems such as motion capture, inertial measurement units (IMUs) have attracted increasing attention as low-cost, portable alternatives. This study aimed to validate a low-cost IMU for measuring barbell velocity during the concentric phase of the deadlift, using a motion capture system as the reference standard.
Methods Sixteen young men (mean age = 21.7 ± 3.4 years) with at least one year of resistance training experience participated. After a standardized warm-up, each participant performed three sets of five deadlifts at 50% of their body mass. Barbell velocity was recorded simultaneously using a triaxial IMU and a 3D motion capture system. Data were analyzed under both filtered and unfiltered conditions. Vertical velocity was derived through sensor fusion and zero-velocity update algorithms. The third and fourth repetitions of each set were included in the analysis. Differences between measurement tools were examined using repeated-measures ANOVA, along with agreement and reliability indices.
Results For mean velocity, no significant differences were found between the IMU and motion capture across processing conditions. The mean bias between devices was small (≈0.02 m/s), with high reliability and agreement indices (ICC ≈ 0.86; RMSE ≈ 0.16 m/s; CV ≈ 14%). In contrast, for peak velocity, significant main effects of measurement tool, filtering, and their interaction were observed. The IMU systematically underestimated peak velocity compared with motion capture, showing larger mean biases (up to −0.19 m/s), moderate correlations, and higher error indices.
Conclusion The IMU demonstrated high agreement and low error for mean velocity, but greater variability for peak velocity. Thus, this low-cost, portable sensor can serve as a practical tool for monitoring mean concentric velocity during deadlifts in field settings, although caution is warranted when interpreting peak velocity values.