The Study Of Triboelectric Motion Sensor And Its Application In Human Physical Activity Recognition

With the increasing development of the wireless body sensors network and wearable devices,human physical activity recognition is widely used in medical diagnosis,well-being management and rehabilitation treatment.The main technology for human physical activity recognition is based on inertial measurement unit including accelerometers,gyroscopes,and so on.These sensors are manufactured very tiny and the collected data are transmitted wirelessly to lower the influence on people's daily life.In the wearable environment,the energy supply for these sensors is typically achieved using batteries,which,however,can not store a large number of energy subject to the limited size.Therefore,the energy consumption for this inertial measurement unit is a great challenge of their application in long term monitoring and recognition tasks.This paper presents a new design and experimental validation of a triboelectric motion sensor in wearable body sensor network for human physical activity recognition and for energy harvesting.The proposed sensing system features dual functions including both motion detection and energy harvesting at very low cost.Compared with traditional motion sensors,this newly designed wearable sensing system is based on triboelectrification and can collect motion signals caused by physical activities without any power supply.In addition,the triboelectric structure can also be used as an energy harvester for motion harvesting due to its high output voltage in random low-frequency motion and a relatively stable voltage when involving continuous activities.Experiments are conducted to validate the activity recognition of this motion sensor by collecting five common activity data: sitting and standing,walking,climbing upstairs,downstairs,and running.The method of K nearest neighboring(kNN),support vector machine(SVM)and logistic regression(LR)are selected to recognize these activities.The results show that our system can perform physical activity recognition with an average successful accuracy of over 80% for walking,sitting and standing and running using this triboelectric motion sensor.In the energy harvesting experiment,a standard energy harvesting circuit with a capacitor is used for validation and calibration of energy harvesting.First,the output voltage is analyzed and simulated theoretically and tested practically.Then the energy from running of a subject is harvested by the triboelectric motion sensor.The results show that the peak power can be achieved up to 9.4 ?W in a one-time motion.The two experiments above validate that the triboelectric motion sensor can be used in human physical activity recognition and have the function for energy harvesting.