A two-step refraction correction model for slope monitoring by measuring robots
Ensuring the precision of robotic total stations in automated slope deformation monitoring is crucial for reliable results.Atmospheric refraction significantly affects this precision.Various refraction correction models aim to mitigate its impact,but common models often introduce meteorological representativeness errors or neglect the spatial position relationships of calibration points used for correction.This paper proposes a two-step refraction correction model comprising distance and coordinate corrections.Initially,meteorological data from the station point's meteorological station is employed for the first correction step:meteorological adjustments are made to slope distance observations,and these corrected distances are used to compute monitoring point coordinates.Subsequently,a second step involves coordinate differential correction at each monitoring point to reduce the influence of meteorological factors on angular measurements.A specific number of calibration points is designated for this correction,adhering to the principle of proximity.Final corrected coordinates are obtained by selecting suitable calibration points for each monitoring point's coordinate differential correction.Experimental analysis based on data from a slope monitoring project using a robotic total station investigates the effect of varying numbers of calibration points on the correction outcome.Results indicate that while increasing the number of calibration points improves the correction,beyond a certain point,additional calibration points do not significantly enhance precision.The correction effect depends not only on the quantity of calibration points but also on the correlation between their atmospheric refraction and that of the monitoring point.For practical verification,it is generally recommended to use the two nearest calibration points for coordinate differential correction.Comparative analysis with typical meteorological correction and distance differential correction methods reveals negligible differences in the mean square errors of monitoring points along the X and Y axes across different refraction correction models.However,the proposed two-step model shows superiority in improving Z-axis precision due to its consideration of vertical angle refraction corrections,which are highly sensitive to meteorological conditions.This study offers valuable insights for selecting an appropriate atmospheric refraction correction model in robotic total station slope monitoring tasks.
slope monitoringrobotic total stationatmospheric refractionrefraction correctioncalibration pointdifferential modelengineering survey
NETL
NSTL
万方数据
