Design And Application Of Soft Capacitance-based Human Motion Estimation System

Monitoring the flexible movement of human limbs in unstructured environments is a challenging task.It is a key issue to study the interaction between human and prosthetic limbs,exoskeletons and other robots.Traditional motion measurement methods are limited in conforming to compliant human joint and adapting to unpredictable external disturbances.Motivated by the need to capture rotation angles of lower-limb joints in real time,this thesis proposes a soft wearable sensing system based on a network of soft capacitive sensors realizing the flexible measurement of human motion in amphibious complex environment.The main contents of this thesis include:The thesis firstly designed the hardware of the measurement system: the AD7746 capacitance measurement chip was selected according to the measurement principle of the flexible strain sensor,and the capacitance measurement and transmission system based on CAN bus was developed.Then the software part of the measurement system was completed: mainly including lower computer programs like the AD7746 driver and range expansion algorithm,and host computer program based on C# which is capable of reading the capacitance signal and realtime drawing.Then,the application of the system in the measurement of lower limb motion of human body is performed: the capacitance-joint angle model is established by using Vicon,and the absolute error of each joint of the lower limb can be lower than 16 degrees;based on the obtained capacitance-joint model and Unity3 D game engine,a visualization program for real-time simulation of human lower limb motion that can be used in the AR/VR field is developed.Finally,a flexible measurement system is applied in the field of human motion intent recognition: using logistic regression algorithms to perform motion modes recognition on the land(flat,stair,stair,and squat)and underwater(walking,freestyle,and breaststroke).The recognition rate was over 97%.