Research On The Algorithms Of Measuring Force Lines Based On Six-axis Inertial Sensor

With the rapid development of science and technology,the lower limb force line measurement system has gradually become a hot spot in the development of the medical industry.As a key technology of the lower limb force line measurement system,indoor positioning technology is considered to be one of the most promising information technologies in the new century.The inertial sensors are widely used due to their small size,portability,low cost and strong autonomy.However,the indoor positioning technology based on inertial sensors is prone to accumulative errors and computational complexity.Therefore,reducing accumulative errors and reducing computational complexity is the main research content of this thesis.This thesis first introduces the research background and significance of the lower limb force line measurement system,expounds the research status of indoor positioning technology at home and abroad,and describes the principles of inertial sensors,data fusion,attitude solution and convolutional neural network.Then,in view of the positioning problem in the lower limb force line measurement system,based on inertial sensors,a series of research work were carried out on how to improve the accuracy of indoor positioning,reduce the complexity of indoor positioning,and improve efficiency.The main contributions of this thesis include the following aspects:1)According to the principle of inertial sensor,a framework of force line measurement system based on inertial sensor is proposed.The framework consists of three parts: data fusion,attitude solution and position solution.In the data fusion part,the accelerometer and gyroscope data are collected by inertial sensor,and then the data are fused to obtain quaternion data.In the attitude solution part,the quaternion method is used to transform the acceleration data in the carrier coordinate system to obtain the acceleration data in the geographic coordinate system.In the position solution part,the indoor position information is obtained by solving the acceleration data in the transformed geographical coordinate system.2)Based on the above research contents,an indoor positioning algorithm based on double integral is proposed.Its main idea is to obtain the position information by double integral of acceleration.In order to prevent the increase of position error caused by double integral,the zero-velocity update algorithm is used to reduce the accumulative error,and on this basis,the zero-velocity update algorithm is improved to further reduce the accumulative error.3)Based on the above research contents,an indoor positioning algorithm based on convolutional neural network is proposed.Its main idea is to normalize the collected data,use the obtained acceleration data and the real position information of the object as training samples,train the convolutional neural network model,and finally predict the position information.The experiment based on the actual sensor data shows that the optimized zero-velocity update algorithm can improve the indoor positioning accuracy more than the zero-velocity update algorithm.The introduction of the convolutional neural network model is more accurate and more stable than using the BP(Back Propagation)neural network model.