Mixed kernel support vector machine gyroscope temperature drift compensation method based on BAS-GSA optimization

Abstract

The invention discloses a mixed kernel support vector machine gyroscope temperature drift compensation methodbased on BAS-GSA optimization. Amixed kernel function based on a linear kernel and a radial basis function kernel is firstly established, so that the mixed kernel function can have relatively strong extrapolation and interpolation capabilities at the same time, and the learning capability and generalization capability of a support vector machine can be effectively improved; and aiming at the defects that a conventional hyper-parameter optimization algorithm GSA is easy to fall into a local minimum value and the local search capability of a BAS is relatively weak, a hybrid BAS-GSA search algorithm with global search capability and local search capability at the same time is established, and hyper-parameter high-precision optimization of a hybrid kernel support vector machine (HKSVM) is realized. The method is verified through a fiber-optic gyroscope temperature drift experiment, and the obtained effect can meet the high-precision temperature drift compensation requirement.

Description

technical field [0001] The invention belongs to the field of inertial navigation devices, and in particular relates to a hybrid kernel support vector machine gyro temperature drift compensation method optimized based on BAS-GSA (beetle whisker search algorithm-gravity search algorithm), aiming at large temperature drift of optical fiber and micromechanical gyro, Features such as strong nonlinearity, realize high-precision temperature drift compensation of gyro. Background technique [0002] The strapdown inertial navigation system is an autonomous and full-information navigation system, and the gyroscope is the key device to realize high-precision navigation and positioning. However, measurement errors caused by temperature drift severely limit the practical application of gyroscopes in navigation and positioning. Since the characteristics of some components are largely affected by the ambient temperature, the resulting gyroscopic measurement errors will vary with temperatu...

AI Technical Summary

Problems solved by technology

However, heuristic algorithms have their own advantages and disadvantages, and there is no method with the best search ability that can combine the advantages of multiple algorithms.

[0004] Improving the temperature drift compensation accuracy of the gyro can also be improved on the hardware, such as adding an auxiliary gyro temperature adjustment device to keep the gyro temperature constant when using the device. This solution can fundamentally solve the temperature drift problem of the gyro device, but makes the device The structure becomes extremely complex, while increasing the device volume and cost, which is not conducive to popularization and application in practice

Images