Research On Evaluation Of Riding Effect Based On Pavement Vibration And Photoelectric Volume Pulse Wave

Cycling has been celebrated as a healthy lifestyle and activity in recent years.To evaluate the effects of cycling and help cyclists individualize an effective workout plan,a portable,multi-parameter detection system is designed by combining an accelerometer with a pulse oximeter,to collect the pavement vibration signals and the original human-body pulse signals during bike riding.Based on these signals,two kinds of parameters are calculated: the pavement roughness index and cyclists' physiological parameters,including pulse rate,blood oxygen saturation,and respiratory rate.The pavement roughness index is proposed to be an effective indicator for reflecting the riding comfort and the cyclists' physiological parameters are used to estimate the effects of exercise.Finally,a support vector machine(SVM)classifier is employed to firstly classify and evaluate the effects of cycling systematically.Below is an overview of the efforts we've made.Firstly,we conduct an in-depth review of the state-of-the-art technologies for measuring pavement roughness and human-body physiological parameters.Based on the quarter car model,we establish a simplified bicycle model which takes the bicycle as a sensor platform to collect pavement vibration data and calculate the International Roughness Index.We also illustrate the principles of deriving pulse rates,blood oxygen saturation,and respiratory rates from photoelectric pulse waves,providing a solid theoretical basis for the experiment and the result analysis.Secondly,we detail the hardware-and-software design of the multi-parameter detection system.We map out three distinct schemes for the cycling experiment and explain how to set up road materials and choose cycling routes in detail.In the experiment,we collect the pavement vibration acceleration data from which the IRI values of various pavements are obtained.Based on these measurements,we map Yanshan University with pavement roughness on its main campus roads.We also collect riders' physiological parameters on a concrete pavement with low roughness and then perform a quantitative analysis to assess the impact of cycling conditions on physiological parameters.Finally,we firstly finish to classify and evaluate the cycling effect based on our experimental data.Our method calculates three features indicating cycling comfort and the impact on cycling exercise: International Roughness Index,Power Spectral Density,and distribution entropy;and gathers three physiological parameters indicating the exercise effects: pulse rate,blood oxygen saturation,and respiratory rate.Taking those six parameters as input features,an SVM classifier is adopted for accurate classification and quantitative evaluation on the effects of exercise under certain cycling conditions.The classification result is helpful to make reasonable suggestions on quantifying a cycling plan.The fruit of this paper can be used to classify the exercise effects of cycling,providing a scientific training basis for cycling enthusiasts and athletes.It can also assist in the development of trip navigation software for urban non-motor vehicles to ensure raised urban bicycle traveling rate and reduced air pollution,which is beneficial both economically and socially.