Research on self-calibration algorithm of micro-inertial sensors for unmanned aerial vehicles

During the flight of an unmanned aerial vehicle, we usually use micro-inertial sensors to collect flight data, so as to obtain the flight attitude and ensure its stable flight. However, the measured values output by the sensors will have errors due to the influence of factors such as temperature and time drift. To improve the accuracy and stability of the sensors, calibration is required. The traditional static multi-position calibration method cannot calibrate all the key parameters of micro-inertial sensors. To address this issue, this paper proposes a calibration algorithm based on least squares fitting. This algorithm establishes error models for micro-accelerometers and micro-gyroscopes, and uses least squares fitting to determine their error compensation coefficients, thereby realizing the self-calibration of the zero-bias errors and non-orthogonal errors of micro-accelerometers and micro-gyroscopes. Through simulations, the residual error of the micro-accelerometer after calibration can be stably controlled within ±2 × 10⁻³. For the micro-gyroscope after calibration, the three axes all show good consistency in zero-bias error calibration, achieving favorable results.