Temperature Compensation Method of Accelerometer Based on the Mixture Kernel Functions Support Vector Machine

Due to the requirements of engineering practice, inertial navigation system should operate over a wider temperature range, and the temperature will cause the output of inertial devices to drift. To minish the impact of temperature on the measurement accuracy of accelerometers in inertial navigation system, the support vector machine is used to fit and regress the scaling factor and bias of the accelerometer's temperature compensation model, and the Harris hawks optimization is applied to search for the optimal parameters of support vector machine and improve the performance of the support vector machine. Finally, the method was validated through physical temperature experiments on the accelerometer. The testing results indicated that the method raised in this paper realized effective temperature compensation. The output stability of the accelerometer is improved by an order of magnitude, which means that the method has practical engineering value.