在煤矿井下位置指纹匹配定位过程中,逐点采集位置指纹需要耗费大量人力且效率较低.提出基于改进Kriging插值的煤矿井下定位算法,只需采集部分位置指纹,通过Kriging插值估计其它位置指纹,通过分组和差分进化策略改进灰狼优化算法,优化变异函数参数选取,提高插值估计精度.实验结果表明,在减少采集工作量目标的同时,相对于IDW插值和普通Kriging插值算法,利用改进的Kriging插值定位算法可进一步提高位置指纹的估计精度和井下定位精度.
In order to solve the problem of collecting fingerprints point by point in the fingerprint matching and positioning process of coal mine underground position, which takes a lot of manpower and low efficiency.This paper proposes a coal mine underground positioning algorithm based on improved Kriging interpolation.It only needs to collect partial position fingerprints and interpolate through Kriging.The fingerprints of other locations are estimated, and the grey wolf optimization algorithm is improved by grouping and differential evolution strategy, the variation function parameters are optimized, and the interpolation estimation accuracy is improved.The experimental results show that compared with IDW interpolation and ordinary Kriging interpolation algorithm, the improved Kriging interpolation algorithm is used to further improve the position fingerprint estimation accuracy and downhole positioning accuracy.