Precise Astronomical Orientation Using Multiple Stars at Their Culmination and Analysis of Different Regression ModelsSource: Journal of Surveying Engineering:;2020:;Volume ( 146 ):;issue: 003DOI: 10.1061/(ASCE)SU.1943-5428.0000323Publisher: ASCE
Abstract: Precise determination of the astronomical azimuth using the local hour angle (LHA) of Polaris is only applicable in northern middle latitudes. Therefore, to achieve fast and precise astronomical orientation in any part of the world, this study proposes continuous observations from the same station of multiple north stars and south stars at their culminating positions, thereby reducing the impact of errors in observation station longitude and time measurement. This method, which uses the regression model, determines the astronomical azimuth from the observation station to the ground target. Model validity is affected by various factors including sample size, data distribution, and data quality; therefore, suitable model conditions and individual characteristics of parametric and nonparametric regression are analyzed. Then, a two-step regression method comprising data binning followed by linear regression is proposed. Experimental data are subjected to direct linear regression, the summation of north and south star data followed by linear regression, nonparametric regression of summed data, and two-step regression of summed data. The two-step regression method produces the most accurate results that agree with those of the LHA method within a range of ±1.2120 × 10−6 rad (0.25″). Compared to the LHA method, the proposed method extends the geographical application range and eliminates the need for precise astronomical positioning prior to determining the orientation.
|
Collections
Show full item record
contributor author | Xinjiang Liu | |
contributor author | Chonghui Li | |
contributor author | Yong Zheng | |
contributor author | Yinhu Zhan | |
contributor author | Chao Zhang | |
date accessioned | 2022-01-30T21:10:09Z | |
date available | 2022-01-30T21:10:09Z | |
date issued | 8/1/2020 12:00:00 AM | |
identifier other | %28ASCE%29SU.1943-5428.0000323.pdf | |
identifier uri | http://yetl.yabesh.ir/yetl1/handle/yetl/4267763 | |
description abstract | Precise determination of the astronomical azimuth using the local hour angle (LHA) of Polaris is only applicable in northern middle latitudes. Therefore, to achieve fast and precise astronomical orientation in any part of the world, this study proposes continuous observations from the same station of multiple north stars and south stars at their culminating positions, thereby reducing the impact of errors in observation station longitude and time measurement. This method, which uses the regression model, determines the astronomical azimuth from the observation station to the ground target. Model validity is affected by various factors including sample size, data distribution, and data quality; therefore, suitable model conditions and individual characteristics of parametric and nonparametric regression are analyzed. Then, a two-step regression method comprising data binning followed by linear regression is proposed. Experimental data are subjected to direct linear regression, the summation of north and south star data followed by linear regression, nonparametric regression of summed data, and two-step regression of summed data. The two-step regression method produces the most accurate results that agree with those of the LHA method within a range of ±1.2120 × 10−6 rad (0.25″). Compared to the LHA method, the proposed method extends the geographical application range and eliminates the need for precise astronomical positioning prior to determining the orientation. | |
publisher | ASCE | |
title | Precise Astronomical Orientation Using Multiple Stars at Their Culmination and Analysis of Different Regression Models | |
type | Journal Paper | |
journal volume | 146 | |
journal issue | 3 | |
journal title | Journal of Surveying Engineering | |
identifier doi | 10.1061/(ASCE)SU.1943-5428.0000323 | |
page | 8 | |
tree | Journal of Surveying Engineering:;2020:;Volume ( 146 ):;issue: 003 | |
contenttype | Fulltext |