Embeded smart sensor: System calibration and modeling

This paper studied the calibration method for a type of embedded smart sensors which are integrated with inertial, magnetic and vision sensors. A set of calibration and modeling methods which are suitable for the low cost applications have been proposed. For the deterministic items and nondeterministic items, we used the method of modulus matching of the input vectors and the method of time series analysis to calibrate and modeling them respectively; for the relative rotation between the accelerometers' frame and the camera frame, we proposed a calibrating method which is based on the gravity vector and plumb line. The experimental and simulation results demonstrated the practicability and effectiveness of these methods.