11:00 〜 11:15
[MSD40-02] Evaluating the Target Pointing Accuracy of the Diwata-2 Using High Precision Telescope Images
★Invited Papers
キーワード：Microsatellite, cross-track error, target pointing, operations
Target pointing accuracy is a crucial factor in capturing satellite imagery. For Diwata-2, a 50 kg microsatellite, the ability to capture a target within the frame of its High Precision Telescope (HPT) depends heavily on its pointing accuracy. Given the 3 km x 2 km frame of an HPT image, just 0.2 degrees of pointing error may result in a capture that is completely off-target.
Over time, the accuracy of a satellite may degrade due to sensor misalignment, orbital decay, and mechanical wear. Hence, regular calibration is necessary to both maintain and improve the satellite’s pointing accuracy. Since its launch in 2018, Diwata-2 has undergone several target pointing calibrations. The latest procedure, completed in June 2020, was able to improve the satellite’s existing target pointing accuracy to <0.66 degrees (roughly 6 km of cross-track distance from target), with 24% of captures falling within <0.1 degrees away from the target (roughly 1km) [1].
In this study, we perform an extended evaluation of Diwata-2’s target pointing accuracy under the abovementioned calibration procedures. The accuracy of the resulting images are measured in terms of target pointing error, which is defined as the cross-track angle or distance between the captured location (image center) and the intended target.
Over 120 HPT payload acquisitions were taken between March and October 2021 at capture angles ranging from 0-7.5 degrees of roll. This range corresponds to ground distances of 0-77 km, which is the practical target range for Diwata-2 captures. After geolocating the images, cross-track errors were estimated using spherical geometry and validated using Google Earth Pro.
Results show that the current cross-track error observed in Diwata-2 images ranges between 0-0.95 degrees. This corresponds roughly to 0-10 km of distance between image and intended target. Out of the 124 captures, 20.2% were found to be <0.1° away from the target as compared to the previously determined 24%. Meanwhile, 38.9% of HPT images were found to capture the target within frame (<1.5 km from target).
Capture angle appears not to have any significant effect on the magnitude of target pointing error, as correlation between cross-track error and capture angle is found to be weak (r=-0.17). Instead, cross-track error is seen to be normally distributed around a mean value of -0.26 degrees. This corresponds to a mean distance of 2.63 km west of the intended target. The westward bias of the mean cross-track error is consistent with our findings wherein 87% of the captured images were seen to fall to the left of the intended target.
The tendency of Diwata-2 to point slightly westward suggests that a uniform offset in the opposite direction may be applied to the capture angle to counter this directional bias. By setting the capture angle offset to +0.26 degrees (equal but opposite to the mean angular cross-track error), the mean cross-track error may shift towards zero, which in turn may potentially improve the satellite’s overall pointing accuracy.
[1] Violan, E., Fujita, S., Sato, Y., Sakamoto, Y., Banatao, J., Kuwahara, T., Yoshida, K., (2021, March). In-Flight Target Pointing Calibration of the Diwata-2 Earth Observation Microsatellite. Paper presented at AeroConf 2021 IEEE Aerospace Conference, Montana, USA
Over time, the accuracy of a satellite may degrade due to sensor misalignment, orbital decay, and mechanical wear. Hence, regular calibration is necessary to both maintain and improve the satellite’s pointing accuracy. Since its launch in 2018, Diwata-2 has undergone several target pointing calibrations. The latest procedure, completed in June 2020, was able to improve the satellite’s existing target pointing accuracy to <0.66 degrees (roughly 6 km of cross-track distance from target), with 24% of captures falling within <0.1 degrees away from the target (roughly 1km) [1].
In this study, we perform an extended evaluation of Diwata-2’s target pointing accuracy under the abovementioned calibration procedures. The accuracy of the resulting images are measured in terms of target pointing error, which is defined as the cross-track angle or distance between the captured location (image center) and the intended target.
Over 120 HPT payload acquisitions were taken between March and October 2021 at capture angles ranging from 0-7.5 degrees of roll. This range corresponds to ground distances of 0-77 km, which is the practical target range for Diwata-2 captures. After geolocating the images, cross-track errors were estimated using spherical geometry and validated using Google Earth Pro.
Results show that the current cross-track error observed in Diwata-2 images ranges between 0-0.95 degrees. This corresponds roughly to 0-10 km of distance between image and intended target. Out of the 124 captures, 20.2% were found to be <0.1° away from the target as compared to the previously determined 24%. Meanwhile, 38.9% of HPT images were found to capture the target within frame (<1.5 km from target).
Capture angle appears not to have any significant effect on the magnitude of target pointing error, as correlation between cross-track error and capture angle is found to be weak (r=-0.17). Instead, cross-track error is seen to be normally distributed around a mean value of -0.26 degrees. This corresponds to a mean distance of 2.63 km west of the intended target. The westward bias of the mean cross-track error is consistent with our findings wherein 87% of the captured images were seen to fall to the left of the intended target.
The tendency of Diwata-2 to point slightly westward suggests that a uniform offset in the opposite direction may be applied to the capture angle to counter this directional bias. By setting the capture angle offset to +0.26 degrees (equal but opposite to the mean angular cross-track error), the mean cross-track error may shift towards zero, which in turn may potentially improve the satellite’s overall pointing accuracy.
[1] Violan, E., Fujita, S., Sato, Y., Sakamoto, Y., Banatao, J., Kuwahara, T., Yoshida, K., (2021, March). In-Flight Target Pointing Calibration of the Diwata-2 Earth Observation Microsatellite. Paper presented at AeroConf 2021 IEEE Aerospace Conference, Montana, USA