Muhammad Uzair, Ali F. Khan, Khawar Khurshid, and Byeungwoo Jeon
Measurement, sensor error model, accelerometer, calibration, robotics
In this paper, we propose a method for performing static calibration of a tri-axial microelectromechanical systems (MEMS) accelerometer utilizing an improved calibration method exploiting a modified sensor error model (SEM). Our key idea is to make calibration parameters a function of orientation angles rather than determining finite discrete calibration parameters. Experimental verification involves putting the SEM to test by attempting to calibrate a printed circuit board (PCB) mounted MEMS tri-axial accelerometer utilizing a laboratory grade gyroscope stabilization platform. Experimental procedures are carried out to estimate the parameters of proposed SEM, such as scale factors, offsets and cross-axis sensitivity, etc. These parameters are subsequently employed to apply real-time corrections to data generated by the accelerometer to determine absolute angular error. Experimental results show that by using the modified SEM, we obtain superior performance than by using similar SEMs thus providing an alternative solution to static calibration of a low-cost MEMS tri-axial accelerometer.
Important Links:
Go Back
