MIMU error calibration method of turntable free platform based on improved genetic algorithm

Micro Inertial Measurement Unit (MIMU) is the core component of the micro inertial navigation system. To obtain motion parameters such as the carrier’s high-precision attitude, speed, and position, MIMU often needs to be calibrated and compensated. The traditional high-precision turntable MIMU calibration is complex and the test equipment is expensive. Aiming at the need of low cost and high precision field calibration of MIMU, an improved genetic algorithm was proposed for MIMU error calibration for turntable free. Firstly, the random noise in the MIMU measurement output is pretreated by wavelet denoising method, then the error function is constructed by designing the MIMU multi-position rotation scheme to stimulate the correlation error, and the genetic algorithm is introduced with the error function as the optimization objective function. Finally, the calibration error parameters in the objective function are searched and optimized to identify the deterministic error of the gyroscope and accelerometer in the MIMU, so as to achieve the field calibration of MIMU without relying on other test equipment. Simulation results show that the proposed algorithm has a high MIMU calibration accuracy, with a calibration error of about 4%.