A research on calibration of low-precision MEMS inertial sensors

Micro-electromechanical systems (MEMS) are an emerging technology that has the potential for a multitude of uses. MEMS sensors are cost-effective units with small dimensions, but suffer from large random errors compared with their mechanical predecessors. It is hard to have an accurate description of sensor errors because the characteristics of errors are quite different among the MEMS inertial sensors with the same specifications. Installation errors result from the imperfection in welding the MEMS sensor which consists of a triaxial accelerometer and a triaxial gyroscope on a single-chip. Low-precision MEMS inertial sensors exhibit biases, scale factors, and large installation errors. According to the error characteristics of MEMS inertial sensors, a mathematical model was analyzed. The validity of the six-position test was verified by using the correction between the triaxial accelerometer and the triaxial gyroscope on one single-chip. One method to identify the intersection angle among the sensitive axes and the measured vector was analyzed. The test results show that the calibration can effectively correct the bias, scale factor and installation error of MEMS inertial sensors.