The Research On Posture Measurement System Based On MEMS

Real-time human motion measurement which is also known as action capture,is widely used in the biomedical,sports,man-machine control,virtual reality,film and television production and other fields.The existing technology to realize the action capture is divided into optical motion capture,magnetic motion capture,mechanical motion capture and inertial motion capture.The optics-based motion capture system is the most widely used,but it is gradually replaced by a more simple and less expensive inertial motion measurement system because of its complex installation and expensive price.The inertial motion measurement system uses a new generation micro-electromechanical sensors to overcome the expensive,jitter,delayed interference,and limited motion range problems in other sensing devices.The main work of this paper are as follows:(1)The building of hardware and software platform of motion capture based on MEMS sensors and wireless WIFI technologyAccording to the requirement of real-time posture measurement,the nine-axis inertia node with three-axis accelerometer,three-axis gyroscope and three-axis Magneto resistance meter is designed.The CAN bus and wireless WIFI communication technology are applied to achieve the high-speed transmission of data,so Inertial node unit data can be acquired in real time.The hardware and software platform of MEMS-based posture measurement system is completed,and 3D virtual human model is used to track the real-time posture and location of the device wearer.(2)Study on the data fusion algorithm to achieve real-time posture solutionA variety of data fusion algorithms are designed and implemented.The complementary filtering and seventh-order extended Kalman algorithm are improved to solve the linear acceleration interference in human motion measurement.The improved algorithm is proved by uniaxial rotation simulation experiment.in addition,an extended Kalman filter algorithm is designed for the standing foot characteristics of human body measurement.The velocity and position information of each inertia node are calculated by the coordinates of standing feet and the skeleton posture.The data is used as the measurement data of extended Kalman algorithm and gets good effect in the simulation experiment.Finally,the lever length tracking algorithm based on PI feedback is designed to achieve the insensitivity of the inertia node to the wearing position.The convergence of the algorithm is proved by experiments.(3)Study zero speed point discrimination,posture calibration,vacant and other algorithmsIn order to track the human trajectory,the zero-speed point discriminant algorithm are studied to judge the foot whether is the standing foot or not.In order to realize the insensitivity of inertia node to the wearing angle,the posture calibration algorithm is studied.In order to track the position of the human body when the feet are not standing,the coordinate calculation algorithm is studied in the flight process.