In the last decades, the amount of space debris in Earth orbit, from spent rocket stages to broken satellites, is constantly growing up. This is a relevant issue, since space junk hampers the access road to space and raises collision threats. Some countries have already identified as unavoidable solution the complete stop of future launches of satellites. Hence, several research projects all over the world are trying to address the space debris removal problem. In our work, we focused on the problem of developing strategies for the non cooperative rendezvous and capture of space debris by exploiting a 3D stereo camera in order to obtain information about their motion: we need to precisely evaluate the space debris kinematics in order to select and design the best removal techniques. Many research projects have investigated radar or other sensors, well established and adopted on Earth. Actually the clear limits in the space forced us to investigate the adoption of a simple couple of cameras mounted on a stereo frameset: such a vision system is compact, very light, low power demanding and not passive (not sending active radiation to an unknown object with unknown reactions), hence an ideal candidate for space mission payloads. Nevertheless the use of a vision system for this purpose is extremely challenging. Strong sun light in the space environment, the absence of Rayleigh scattering and almost uniform colours of the target highly reduce the accuracy of depth from stereo approaches. The same problems affect the estimation of the kinematics and the evaluation the inertial forces the debris is subject to, which require a reliable tracking of feature points on theobject surface. To this end, we developed a robust approach for obtaining, from a sequence of stereo images, the motion history of clouds of 3D points on the debris surface. The approach is based on first finding robust correspondences between images in a stereo pair and, then, on exploiting these correspondences to obtain a robust registration between consecutive stereo frames.

Space-debris motion reconstruction for removal / Allasia, Walter; Bottino, ANDREA GIUSEPPE. - STAMPA. - (2015), pp. 1-12. (Intervento presentato al convegno XII Conference of the Italian Association of Aeronautics and Astronautics tenutosi a Torino nel 17-19 November, 2015).

Space-debris motion reconstruction for removal

BOTTINO, ANDREA GIUSEPPE
2015

Abstract

In the last decades, the amount of space debris in Earth orbit, from spent rocket stages to broken satellites, is constantly growing up. This is a relevant issue, since space junk hampers the access road to space and raises collision threats. Some countries have already identified as unavoidable solution the complete stop of future launches of satellites. Hence, several research projects all over the world are trying to address the space debris removal problem. In our work, we focused on the problem of developing strategies for the non cooperative rendezvous and capture of space debris by exploiting a 3D stereo camera in order to obtain information about their motion: we need to precisely evaluate the space debris kinematics in order to select and design the best removal techniques. Many research projects have investigated radar or other sensors, well established and adopted on Earth. Actually the clear limits in the space forced us to investigate the adoption of a simple couple of cameras mounted on a stereo frameset: such a vision system is compact, very light, low power demanding and not passive (not sending active radiation to an unknown object with unknown reactions), hence an ideal candidate for space mission payloads. Nevertheless the use of a vision system for this purpose is extremely challenging. Strong sun light in the space environment, the absence of Rayleigh scattering and almost uniform colours of the target highly reduce the accuracy of depth from stereo approaches. The same problems affect the estimation of the kinematics and the evaluation the inertial forces the debris is subject to, which require a reliable tracking of feature points on theobject surface. To this end, we developed a robust approach for obtaining, from a sequence of stereo images, the motion history of clouds of 3D points on the debris surface. The approach is based on first finding robust correspondences between images in a stereo pair and, then, on exploiting these correspondences to obtain a robust registration between consecutive stereo frames.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2621002
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo