基于共轭次梯度算法的磁传感器安装误差校正

Research on Calibrating Installation Error of Magnetic Sensor Based on Conjugated Approximate Gradient Algorithm

随着头盔显示技术的发展,对头盔电磁定位的精度要求越来越高,在现有的加工工艺和机上安装条件下,要提高定位精度,必须对传感器安装误差进行校正。通过坐标转换原理对安装误差进行建模,在此基础上利用最小二乘原则选取目标函数,将校正过程抽象为最优化问题求解。提出一种基于共轭次梯度算法的最优化问题求解方法,实现了传感器安装误差校正。利用MATLAB上对校正算法进行仿真,仿真结果表明,共轭次梯度算法简洁易用、校正精度高、收敛速度快,能够明显提高磁场测量精度。

With the development of HMD technology, in order to improve precision of positioning system it is necessary to calibrate installation error of magnetic sensor under the condition of current processing technic and installation. Coordinate rotation was applied to establish and analysis an error model, and then the goal function was selected using least squares rule based on the model. Calibration of installation error was finally translated into a optimization problem. A kind of method to solve the optimization problem based on conjugated approximate gradient algorithm was put forward to estimate and correct the installa- tion error. The algorithm was simulated in the MATLAB platform. The result indicates that conjugated ap- proximate gradient algorithm can realize easily, achieve high precision, and has fast convergence rate, the algorithm can improve the precision of the magnetic field measurement obviously.