CAJ | 학술논문

针对现阶段地磁传感器误差补偿算法采用的椭球模型假设对输入的校准参数有较高的要求,容易出现较大的拟合误差问题,提出了一种基于模糊逻辑系统的数据分布自判定算法,用来判断校正数据的分布是否合理,使其适当地分布在整个椭球面,准确地拟合出理论的椭球模型. 通过仿真分别求取最小二乘拟合误差和模糊逻辑系统计算误差,结果表明,利用模糊逻辑能够有效的剔除分布不合理的数据样本,解决了校正数据分布不合理带来较大拟合误差的问题,有利于提高地磁传感器的误差自动校正的精度.
침대현계단지자전감기오차보상산법채용적타구모형가설대수입적교준삼수유교고적요구,용역출현교대적의합오차문제,제출료일충기우모호라집계통적수거분포자판정산법,용래판단교정수거적분포시부합리,사기괄당지분포재정개타구면,준학지의합출이론적타구모형. 통과방진분별구취최소이승의합오차화모호라집계통계산오차,결과표명,이용모호라집능구유효적척제분포불합리적수거양본,해결료교정수거분포불합리대래교대의합오차적문제,유리우제고지자전감기적오차자동교정적정도.
Considering that the ellipsoid model assumption used in the conventional error compensation algorithm of magnetometers always has higher demand for the inputted calibration parameter, which easily leads to higher fitting errors. Thus, this paper proposes a data distribution self-judging algorithm based on the fuzzy logic system, which judges whether the distribution of calibration data is reasonable, so as to make them distribute throughout the whole ellipsoid surface properly in order to be accurately fitted to the theoretical ellipsoid model. The errors from the least squares fitting and the calculation of fuzzy logic system were obtained respectively through simulation. The results show that the fuzzy logic system is able to eliminate data samples that are not distributed reasonably, solves the problem of larger fitting error caused by unreasonable distribution of correction data, and is beneficial to improve the accuracy of magnetic sensor error calibration in automatic correction.