부산대학교 도서관

This paper presents the study of Nordic Walking providing objective evaluations based on real time acquisition of kinematic parameters during the sport practice. It is possible to carefully monitor the athletic gesture through the integration of conventional poles with inertial sensors, composed of a triaxial accelerometer, a triaxial gyroscope, a pressure sensor positioned on the handle, and a load cell, which constitute a Wireless Sensor Network whose nodes are appropriately synchronized. The integration of such sensors, which must be unobstructive and not change the functionality of the poles, is dictated by the ultimate goal of providing a real time biofeedback in two possible scenarios. The first is intended for Nordic Walking’s instructors, who have the opportunity to verify the proper practice execution by their trainees through the availability of real time objective data, in addition to their personal experience. The second is devoted to amateur players who can practice alone, after the training session with the instructor, and can independently correct any imperfections in real time using a software tool running on their smartphone. Using the Dynamic Time Warping algorithm, the proposed system identifies the most frequent errors in performing athletic gesture, allowing adjustment in real time of the sporting exercise, through the detection, quantification and correction of errors. The obtained results show that the developed system is able to provide an accurate analysis of the athletic gesture and the proposed algorithm allows a quantitative monitoring of the progress achieved by each subject over time.