Enhanced flyby science with onboard computer vision: Tracking and surface feature detection at small bodies

Abstract

Spacecraft autonomy is crucial to increase the science return of optical remote sensing observations at distant primitive bodies. To date, most small bodies exploration has involved short timescale flybys that execute prescripted data collection sequences. Light time delay means that the spacecraft must operate completely autonomously without direct control from the ground, but in most cases the physical properties and morphologies of prospective targets are unknown before the flyby. Surface features of interest are highly localized, and successful observations must account for geometry and illumination constraints. Under these circumstances onboard computer vision can improve science yield by responding immediately to collected imagery. It can reacquire bad data or identify features of opportunity for additional targeted measurements. We present a comprehensive framework for onboard computer vision for flyby missions at small bodies. We introduce novel algorithms for target tracking, target segmentation, surface feature detection, and anomaly detection. The performance and generalization power are evaluated in detail using expert annotations on data sets from previous encounters with primitive bodies.

Novel 3U Stand-Alone CubeSat Architecture for Autonomous Near Earth Asteroid Fly-By

“…So, it is important that the navigation system (and strategy) is robust to that uncertainties. Many techniques have been investigated or are under current investigation, from more traditional centerof-brightness and geometric centering [11,22], to brand-new deep-learning and neural network based [23,24]. Indeed, the fast growth in neural-networks based navigation will characterize the next years of small bodies exploration.…”

Factorial inferential grid grouping and representativeness analysis for a systematic selection of representative grids

“…Due to this advantage, FIGGRA may be an advanced and reliable alternative algorithm for Earth imagery classification and the associated diverse studies or practices [Jiang and Shekhar, 2017;Rauniyar et al, 2017;Schwenk et al, 2017]. In addition, FIGGRA-based spatial/temporal heterogeneity analysis may facilitate improvement of various quantitative analysis approaches for investigating many problems in Earth and space sciences, e.g., monitoring network design [Mishra et al, 2016;Gleason et al, 2017], urban ecology [Bardhan et al, 2016], representative-days selection [Rife et al, 2013], O 3 distribution detection [Parrish et al, 2016], spatial tracking or navigation [Fuchs et al, 2015;Palmer et al, 2016], tsunamis modeling [Grawe and Makela, 2015], atmospheric process analyses [Weisz et al, 2015], seafloor venting detection [Smart et al, 2017], sporadic E propagation [Ghosh and Berkey, 2015], or eco-system analysis [Zhang et al, 2017]. The potentiality of FIGGRA in addressing the abovementioned problems would be assessed in the future studies.…”

Martian Dust Devil Detection Based on Improved Faster R-CNN

“…Their research automatically detects and tracks plumes on the 67P/Churyumov-Gerasimenko comet using OSIRIS/Rosetta image sequences. Additionally, there have been studies on surface feature detection and tracking on small bodies using onboard computer vision techniques [29], [30]. These studies provide insights for tracking dust devils on other celestial bodies.…”