为了能够实时捕捉人体运动,设计了一套可穿戴式人体运动捕捉系统。它通过分布在人身体上的惯性测量单元来获取人体实时姿态信息,每个惯性测量单元由微型MEMS 3轴陀螺仪、MEMS 3轴加速度计、MEMS 3轴磁力计和MCU组成,MCU获得各传感器数据并利用基于四元数的扩展Kalman滤波解算出对应部位姿态角,通过CAN总线和Bluetooth模块将数据实时上传至计算机,计算机通过VC++和OpenGL程序驱动虚拟人体运动,实现实时人体运动再现。
A wearable human motion capture system is designed in order to be able to capture real-time human motion. It is based on a network of inertial measurement units distributed on the human body to estimate the real time orientation of the limbs. Each of these modules is provided by miniature MEMS 3-axes accelerometer, 3-axes gyroscope, 3-axes magnetometer, and a MCU for collecting measurements from the sensors and computing the orientation through quaternion based extended Kalman Filter. The real time data is uploaded to the computer via the CAN bus and Bluetooth module, computer drives the human body model through VC~+ and OpenGL program to reconstruct body motion.