| Human trunk kinematics research is important to achieve social and economic benefits.Wearable sensing systems are widely used especially in this field,but there are special requirements related to the design and function of these devices.This thesis introduces the development of a wearable sensing system used for human kinematics research.This research builds a human trunk model to track the spine movement and to standardize the experiment method,designs the hardware and software of the wearable system,and develops a sensor fusion algorithm to estimate human trunk sway.Experiments are performed to validate performance of the wearable system and accuracy of the algorithm.By building the human trunk model,this research designs an experiment to study the optimal sensor placement,and normalizes the method to monitor trunk sway.Based on the results,a single sensor should be placed at the spine location T7-T8 for standing tasks(0.54 deg error)while at the spine location T9-T10 for dynamic gait walking tasks(1.02 deg error)to get an accurate estimation of human trunk sway.The results would also support the wearable system calibration.The innovative wearable system is composed of the hardware device and the user interface.The device is packaged with silicon,and the dimensions are 22.5 mm x 20.5 mm x 14.9 mm.It can collect and transmit sensor data wirelessly.The communication distance is 3 meters and battery life is more than one hour.The user interface receives,processes and saves the data,then displays them in real-time.An experiment is performed to validate the accuracy of algorithm which based on Kalman filter.The results show that the error of the algorithm for standing tasks and dynamic gaits are 1.33 deg and 1.81 deg,respectively.The errors satisfy the design requirements of less than 2 deg of error.The thesis shows that the wearable system is well designed and is suitable for practical applications.This system also may make progress in human kinematics research. |
PDF Full Text Request