IMU Hand-Calibration for Low-Cost MEMS Inertial Sensors

This paper proposes a technique to calibrate a low-cost inertial measurement unit (IMU) in the field without relying on expensive laboratory-based equipment. The main parameters to be calibrated in any IMU are misalignment factors, sensitivity scale factors, and gyro and accelerometer biases, all of which are temperature dependent. Usually, IMU manufacturers have highly accurate and expensive rate tables and centrifuges that can provide high-resolution angular rates to the equipment under test but require costly regular maintenance and highly trained technicians. We propose an alternative method of performing this calibration by hand in the field, using a special Kalman filter that benefits from zero velocity measurement updates. The performance of the proposed calibration technique is evaluated and validated via extensive simulations and real experiments.