Abstract
This paper presents IART, a novel inertial wearable system for automatic detection of infringements and analysis of sports performance in race walking. IART algorithms are developed from raw inertial measurements collected by a single sensor located at the bottom of the vertebral column (L5–S1). Two novel parameters are developed to estimate infringements: loss of ground contact time and loss of ground contact step classification; three classic parameters are indeed used to estimate performance: step length ratio, step cadence, and smoothness. From these parameters, five biomechanical indices customized for elite athletes are derived. The experimental protocol consists of four repetitions of a straight path of 300 m on a long-paved road, performed by nine elite athletes. Over a total of 1620 steps (54 sequences of 30 steps each), the average accuracy of correct detection of loss of ground contact events is equal to 99%, whereas the correct classification of the infringement is equal to 87% for each step sequence, with a 92% of acceptable classifications. A great emphasis is dedicated on the user-centered development of IART: an intuitive radar chart representation is indeed developed to provide practical usability and interpretation of IART indices from the athletes, coaches, and referees perspectives. The results of IART, in terms of accuracy of its indices and usability from end-users, are encouraging for its usage as tool to support athletes and coaches in training and referees in real competitions.
Highlights
IntroductionWearable technologies in sports applications are useful tools for measuring the athlete’s performance in outdoor conditions, and they can play a relevant role to support training
Wearable technologies in sports applications are useful tools for measuring the athlete’s performance in outdoor conditions, and they can play a relevant role to support training.as they can provide accurate and reliable data from the athletes, they can be used for developing tools able to support judgments.A sport discipline in which wearable tools can be used for both performance and infringement assessment is race walking, a historical long-distance discipline within the athletics program
We validate our proposal for estimation of LOGC timing (LOGCT) with respect to (1) benchmark values obtained by an high-speed camera system and (2) values obtained by the method proposed by Lee’s method (Lee) et al in [9], which consider a fixed value for the threshold E in Equation (3)
Summary
Wearable technologies in sports applications are useful tools for measuring the athlete’s performance in outdoor conditions, and they can play a relevant role to support training. Authors in [10] presented a method based on machine learning algorithms for identification of race walking infringements (LOGC and bent knee) They started to test a system composed by seven inertial sensors (with sample frequency of 60 Hz) and they involved eight experts Italian race walkers. The overall contribution of the current work are (i) novel method for estimation of LOGC timing based on elite athlete’s kinematics, (ii) novel classification method for LOGC events based on eye limits and World Athletics competition rules, (iii) intuitive radar chart representation of infringements and performance indices, and (iv) validation of the results with a large number of elite athletes. Three appendices are used to provide additional details of the methodology and results
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
