Trademark 3840-480

Monitoring barbell velocity in strength training with IMU to help the scientific and sporting community evaluate lever kinematics

Hello everyone, Inertia Master Technology is dedicated to sharing the cutting-edge scientific research information of IMU, and today we share with you a paper to verify the validity and reliability of IMU (Inertial Measurement Unit) to automatically acquire barbell speed during bench press.

To verify the validity and reliability of an IMU-based device to automatically acquire barbell velocity and key performance indicators during bench press exercise, the Lorenzo Rum, Italy, scientific team assembled seven Paralympic weightlifters (one female, six male) and had them perform four maximal bench presses after warming up while using a high-resolution, high-speed video camera (Hero 9 GoPro) and IMU (MTw, Xsens) to record the trajectory of the barbell during the athletes' bench press.

Fig. 1 Picture of the experimental setup White circle (first from left) shows the position of the reflection marker White circle (second from left) shows the position of the IMU

The camera was fixed on a tripod, perpendicular to the longitudinal axis of the barbell, to record the trajectory of the reflective markers placed transversely on the barbell end caps.The IMU device was fixed to the barbell on the same side as the reflective markers, close to the hand grips, but not interfering with the athlete's hand movements.

The method of obtaining barbell velocity profiles from the two measurement systems is as follows:

- Camera data processing:Vertical displacements of the markers obtained from video recordings were low-pass filtered by a second-order Butterworth filter with a cutoff frequency of 20 Hz, and then the first-order derivatives were calculated to obtain barbell velocities.

- IMU data processing:After calibrating and removing the gravitational acceleration, a criterion for the triaxial linear acceleration was calculated. The obtained signal was then low-pass filtered with a second-order Butterworth filter with a cutoff frequency of 10 Hz and numerically integrated to obtain the barbell velocity.

Figure 2 Barbell vertical velocity and typical events for the bench press

The timing of specific events in the sleeper act was automatically identified from the IMU and video data using a previously validated algorithm. Briefly, this customized algorithm automatically detected events based on the identification of peaks, slopes, or thresholds of signal changes, and six events were detected (as shown in Figure 2)

The camera data were used as the reference data for this experiment, and the data obtained through the IMU were statistically analyzed and compared with the camera data to test the validity and reliability of the IMU in recording barbell velocity. The assumptions of normal distribution and homogeneity of data for discrete velocity parameters were verified using Kolmogorov-Smirnoff and Brown-Forsythe, respectively. For all discrete velocity parameters, the agreement between the measurement system of the IMU and the measurement system of the video was evaluated by the Bland-Altman test.

The results show that the IMU data are in good agreement and very similar to the video-based reference data, with a mean difference of 0.00 ± 0.04 m-s between the IMU and the high-speed camera -1. The experimental results show that the IMU data have good reliability.

With the further development of science and technology, IMU is continuously optimized, the error of data will be infinitely close to zero, and can quite accurately capture and record the trajectory of the movement, providing faster, more comprehensive and more accurate data to help professional athletes and fitness enthusiasts to develop a more reasonable and efficient training program, effectively reduce sports injuries and improve sports performance.

Original paper:

Inertia Master Technology--To make quality products

Taobao store 

New stock IMU chips genuine original guarantee