Main content area

A novel approach for estimation of residual attitude of a remote-sensing satellite

Dubey, Bhaskar, Kartikeyan, B.
International journal of remote sensing 2018 v.39 no.22 pp. 8219-8245
nonlinear models, remote sensing, satellites
This article presents a novel approach for estimation of the residual attitude of a remote-sensing satellite based on satellite images with ancillary information and ground control points (GCPs). First, a non-linear model which translates the residual errors in roll, pitch, and yaw to scan-errors and pixel-errors in the image space is established. Subsequently, using the model and given scan-errors and pixel-errors at GCPs, an estimate of residual roll, pitch, and yaw based on the least square minimization of residuals in conjunction with the Newton’s method for non-linear optimization is proposed. A simulation is carried out to show that the estimates of residual roll and pitch are within 0.0008° (equivalent to 0.5 pixel) and the residual yaw is within 0.015° (equivalent to 0.75 pixel at the extreme ends) to its true values. The results of the article can be applied to determine the residual attitude of any remote-sensing satellite. We demonstrate our results by estimating the residual attitude based on the data-products of various Indian remote-sensing satellites. The effectiveness of the approach is shown by comparing the results with that of existing technique and it is concluded that the presented technique estimates residual attitude more accurately than the existing method.