Autonomous Planetary 3-D Reconstruction From Satellite Images

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


    A common task for many deep space missions is autonomous generation of 3-D representations of planetary surfaces onboard unmanned spacecrafts. The basic problem for this class of missions is, that the closed loop time is far too long. The closed loop time is defined as the time from when a human operator detects an interesting object or feature on a, say, asteroid, to a command is issued to aim a science instrument at the feature. This delay may be in the range of hours for all except the objects closest to the Earth. Because the transit time of a typical interesting feature is in the range of seconds to a few minutes, the closed loop time effectively precludes active human control.The only way to circumvent this problem is to build an artificial feature extractor operating autonomously onboard the spacecraft.Different artificial feature extractors are presented and their efficiency is discussed.Based on such features, 3-D representations may be compiled from two or more 2-D satellite images. The main purposes of such a mapping system are extraction of landing sites, objects of scientific interest and general planetary surveying. All data processing is performed autonomously onboard in order to reduce data transfer to ground station and to eliminate the closed loop time delay.
    Original languageEnglish
    Title of host publicationAutonomous Planetary 3-D Reconstruction From Satellite Images
    Publication date1999
    Publication statusPublished - 1999
    EventThe Sixth Internal Workshop on Human Interface Technology - Aizu-Wakamatsu, Japan
    Duration: 1 Jan 1999 → …
    Conference number: 6


    WorkshopThe Sixth Internal Workshop on Human Interface Technology
    Period01/01/1999 → …
    OtherNo exact date found, but held in 1999.

    Cite this