Maximum Likelihood Approach for Low-Cost MEMS Triaxial Accelerometer Calibration

Accelerometer is one of the essential components of Inertial Measurement Units. Due to the issues in manufacturing process and sensor installation, errors are introduced into measurements as results of non-orthogonal error, scale error, zero bias and misalignment. Raw readings without calibration might seriously affect the accuracy of inertial navigation system. In order to obtain accurate attitude estimates it is essential that the accelerometer and inertial sensor axes are aligned and the accelerometer is well calibrated. This paper proposes an off-line calibration method for accelerometer based on Maximum Likelihood Estimation, which compensates for abovementioned errors and determines the alignment between the accelerometer and the inertial sensor axes. The resulting calibration procedure requires no additional equipment which makes the algorithm an easy-to-use one. The efficiency of proposed algorithm is tested by both numerical simulation and low-cost sensor based experiment.