The Research Of Human Motion Capture System Based On The Micro Inertial And Wholly-owned State Measurement

The human motion capture technology based on the micro inertial andwholly-owned state measurement is a kind of technology which is differentform that technology based on optical, acoustic or electromagneticprinciple. Comparing with the traditional motion capture technology, it ishigh real-time, not subject to regional restriction, easy to install, etc.It becomes one of the spot of research in recent years. In recent years, themicro-electromechanical systems (MEMS) technology which has maturedpromotes the vigorous development of the micro-inertial system. It alsomakes the micro inertial sensing element has smaller size, lighter weight,lower power consumption and lower cost. The development of the microinertial system also promotes the continuous of the theory and method. Atthe same time, the multi-sensor data fusion theory and communicationtechnology has improved steadily. The development and advancement ofthe theory and technology have laid a solid foundation for micro inertialattitude measurement-based human motion capture system.On the basis of the summary and analysis of the inertial navigationsystem theory, it puts forward the design method of Micro inertial Attitudeand Heading Reference System (MAHRS) which is applied to humanmotion capture system. And it has derived a specific calculation method.According to the requirements of the human attitude capture system andupdating rate, it puts forward the fourth-order Picard algorithm as theattitude updating solution method. As the same time, it uses the improved Kalman filtering information fusion algorithm based MARG to design awhole attitude measurement device. By the experiments, it shows that thisdevice is full compliance with the requirements of human motion capturesystem. It puts forward the human body model which suit humanmotion capture system based on the micro inertial and wholly-owned statemeasurement. It puts forward the solving methods of the perspective ofhuman motion attitude. And it shows the projections process of the bodyposture. It puts forward the solving methods of position and speed in thishuman motion capture system. It uses an improved mean filter method anddead-processing technology to improve the precision.On the basis of theoretical analysis and algorithm design, embeddedtechnology has been used to complete the human motion capturesystem based on the micro inertial and wholly-owned state measurement.And C language plug-ins which is based on the game engine has beenused. By combining the animation model established by the computer, theoperation and displaying of this system has been realized. The practiceshows that this system is reliable and real robust. And it has reached apreliminary designing requirements of the human motion capture system.It has laid a good foundation for the future of the product.