Research On Intelligent Monitoring Methods And Application Technologies For The Skiing State

The development of racing skiing sports depends on scientific and technological progress,and science and technology must play an essential role in supporting and safeguarding the development.The performance of skiing is affected by various conditions.Effective monitoring,combined analysis,and objective feedback of various motion state parameters are important to improve the skiing skills of athletes.When monitoring skiers moving at high speed in a wide range of venues,there are some problems such as difficulty accurately obtaining the motion state parameters,lack of quantitative evaluation of training effect and timely feedback.It is urgent to establish a comprehensive monitoring and auxiliary training system for the skiing state.This thesis focuses on intelligent monitoring methods and application technologies for the skiing state.Some critical researches have been carried out for the key issues in monitoring theory and system design,signal processing,data analysis,system integration and application.These researches include the theory of skiing sports condition monitoring and intelligent solution,the time synchronization method of multi-source sensing data,the analysis and identification methods of skiing technique action,and the integration and application technology of intelligent auxiliary training system.On this basis,an intelligent auxiliary training system for skiing was built.The system has been applied to the preparation of professional skiing teams for the Winter Olympics and the Winter Paralympic Games,helping athletes improve their performance.(1)The influencing factors of skiing state are studied.In daily skiing training,it is considered necessary to monitor multi-dimensional skiing condition parameters,such as video,kinematic parameters,kinetic parameters,physiological parameters,techniques and tactics.The monitoring requirements of the intelligent monitoring system for real-time monitoring of multiple athletes throughout the whole process,simultaneous analysis of multi-dimensional parameters,and automation of the whole process are analyzed.On this basis,a multi-source data monitoring method integrating wearable sensing,flexible sensing,and multi-angle video acquisition technologies is proposed.Meanwhile,the overall architecture of the intelligent monitoring system and the schemes of the sensing layer,network layer,and application layer scheme are designed.(2)The overall scheme of time synchronization of multi-source sensing data based on satellite timing standard time is designed.In order to solve the problem of automatic synchronization between an independent inertial measurement unit(IMU)without built-in satellite timing module and a global navigation satellite system(GNSS),a multi-source data time synchronization algorithm based on cross-correlation analysis and 0/1 speed dichotomy is proposed.Through static detection,IMU and GNSS data are divided into static interval and motion interval.The speed within the corresponding interval is assigned 0 or 1,respectively,as the inputs of cross-correlation analysis.In the case of changes in the ski area and athletes,it still has the advantages of low computational cost,high precision,and strong robustness.Fully automatic time synchronization between independent IMU and GNSS is achieved when quasi-random human signals are used as input.(3)To realize the analysis of skiing motion characteristics and the intelligent recognition of technique actions,flexible sensors based on triboelectric nanogenerators are designed and optimized.On this basis,a pressure sensing insole and a smart ski pole sensing unit based on the customizable and flexible sensor are fabricated.The effectiveness of the flexible sensor for sensing joint angles and kinetic parameters is verified by combining theoretical analysis,simulation,and experimental studies.A method of analysis and recognition of skiing technique actions is proposed,which combined flexible sensing system,automatic feature extraction algorithm of peak finding,and machine learning.In a wide range of venues,the whole process monitoring of athletes' motion state parameters such as the poling number,stride number and technique action cycle are achieved.The accurate recognition of typical sub-techniques of classical Cross-Country Skiing is realized.(4)Through the comprehensive application of various technologies,the software and hardware integration of the monitoring system is realized.The intelligent auxiliary training system for skiing is developed,which integrates data acquisition,transmission,analysis,and feedback.The effective collection,quantitative evaluation,and real-time feedback of the multi-dimensional parameters(i.e.,physiological,kinematic and kinetic parameters of the whole process)of skiers moving at high speed in a wide range of venues are achieved.The application technology of the system is studied.The system is fully automated,highly applicable and less disruptive,and can be used independently by coaches or athletes without technical support.The system has been used in the daily training of professional skiing teams in Cross-Country Skiing and alpine skiing.