Methodology: Inertial sensors units consisting of triaxial +/-6g accelerometers (x,y,z) and 3 x 500 deg/sec (roll, pitch, yaw) were used. The units sent their data via a class 1 Bluetooth connection to purpose built software on a PC for collection and analysis. The data was analysed using MATLAB software. The protocol consisted of the player performing the two types of throw-in commonly used in soccer; the stationary throw-in, and the running approach throw-in. The inertial units were placed firstly on the chest and then on the sacrum which is an approximation of the Centre Of Mass.
The results from the Centre of Mass inertial sensor unit were able to clear show each footstep in the running approach. The backwards rotation and the forward whipping motion of the torso was also clearly visible. The results from the Chest inertial sensor unit were able to show the torso rotation clearly since the movement was more exaggerated compared to the Centre Of Mass. The footsteps in the run up were not apparent in the Chest data. The Chest data was much noisier than the Centre Of Mass data since the Centre Of Mass data was dampened due to the more limited movement of the body at that position. Conclusion: The inertial sensor units on the Centre Of Mass were better for determining the overall throw-in features since they could clearly show the running approach and were less noisy than the inertial sensor units on the Chest.
Sports Technology >