基于MEMS陀螺仪辅助的粒子群优化磁力计校正

目前磁力计校正中存在需要采集大量数据、获取良好的初值和已知准确的传感器噪声分布等问题，传统的粒子群优化磁力计校正算法能够解决以上问题，但是该算法只能用于磁力计简化模型，校正其中9个误差参数，造成补偿不准确的问题。该算法借助MEMS陀螺仪建立矢量目标函数，采用随机漂移粒子群优化算法估计磁力计12个误差参数，具有较强的全局搜索能力和动态适应性。经过仿真与实测实验表明，该算法在磁力计绕其任意2个单轴不完整旋转1周即可实现校正，并且能够在磁场变化情况下保持精度，相比于传统算法补偿精度高、操作简单。

Particle Swarm Optimization with MEMS Gyro-aided for Magnetometer Calibration

With the particle swarm optimization algorithm, there is no need to collect a large amount of data, obtain good initial value or get accurate sensor noise distribution, which are necessary in magnetometer calibartion. Howerver, the traditional particle swarm optimization algorithm can only be used for the simplified model of the magnetometer to calibrate 9 of the errors, making the compensation inaccurate. In this paper, the vector objective function is established through MEMS gyroscope, and the random drift particle swarm optimization algorithm is used to estimate 12 error parameters of magnetometer, which has high global searching ability and dynamic adaptability. The simulation and actual experimental results show that the calibration can be achieved with the proposed algorithm when magnetometer performs an incomplete rotation around any two axes, and the accuracy can be maintained under the change of magnetic field. Compared with the traditional algorithm, the proposed algorithm has the advantages of high compensation accuracy and simple operation.