摘要 煤矿井下信号存在的非视距传播(Non-line of sight,NLOS),严重影响井下定位精度,提出一种基于遗传神经网络和Taylor算法相结合的煤矿井下定位算法。该算法通过训练遗传神经网络来拟合待测点坐标值,定位基站和参考基站之间的到达时间差(time difference of arrive,TDOA)测量值之间的映射关系,将遗传神经网络得到的定位结果代入Taylor算法,得到最终的定位结果。仿真分析表明,该算法相较于传统的室内定位算法,在非视距环境下具有更好的定位效果。 Aiming at the non-line of sight(NLOS)propagation of underground signals in coal mine,which seriously affects the downhole positioning accuracy,a novel underground positioning algorithm based on the combination of genetic neural network and Taylor algorithm is proposed.This algorithm trains the genetic neural network to fit the mapping relationship between the coordinate value of the measurement point and the time difference of arrival(TDOA)measured value between the location base station and the reference base station,and then puts the positioning results obtained by the genetic neural network into the Taylor algorithm to obtain the final positioning results.Simulation results show that compared with traditional indoor positioning algorithm,this algorithm has better positioning effect in non-line-of-sight environment.
出处 《实验室研究与探索》 CAS 北大核心 2021年第4期8-12,共5页 Research and Exploration In Laboratory
基金 国家自然科学基金项目(61503224) 山东省自然基金项目(ZR2017MF048) 山东省研究生教育质量提升计划建设项目(2016050) 青岛市民生科技计划项目(17-3-3-88-nsh)。
关键词 遗传算法 神经网络 TDOA Taylor算法 genetic algorithm neural network time difference of arrive(TDOA) Taylor algorithm
