Abstract—We developed a new motion recording system based on GPS and 9 Degrees of Freedom sensors. Deploying the system, we analyzed turn maneuvers on a halfpipe with skateboard, inline skate, and BMX. The recording data was processed by pre-processing software that synchronized the GPS timestamp on the data collected from the multiple devices. Based on the pre-processed data, we defined a distance function between each pair of time series of two runs in order to identify the varieties of turn maneuvers on a halfpipe. For comparison, we also calculated an ordinal distance measure based on basic statistical indexes. In order to evaluate the two distances measured, we defined a precision of ranking for each distance. Furthermore, to understand similarities among runs, we illustrated two dendrograms of hierarchical clustering using both the proposed method and the ordinal method. In terms of the precision score and dendrogram, we confirmed that the proposed measure is better than the ordinal measure to describe differences in ride data on a halfpipe. 
 

Index Terms—9 degrees of freedom sensor, action sports, BMX, GPS, halfpipe, inline skate, skateboard

Cite: Kazunari Ishida, "Classification of Turn Maneuvers on Halfpipe with Multiple Sensors," International Journal of Electronics and Electrical Engineering, Vol. 4, No. 5, pp. 420-425, October 2016. doi: 10.18178/ijeee.4.5.420-425