JPL Technical Report Server

A Minimal State Augmentation Algorithm for Vision-Based Navigation without Using Mapped Landmarks

Show simple item record San Martin, Miguel A. Bayard, David S. Conway, Dylan T. Mandic, Milan Bailey, Erik S. 2019-08-21T16:46:12Z 2019-08-21T16:46:12Z 2017-05-29
dc.identifier.citation 10th ESA GNC Conference, Salzburg, Austria, May 29-June 2, 2017 en_US
dc.identifier.clearanceno CL#17-2410
dc.description.abstract This paper describes MAVeN (Minimal State Augmentation Algorithm for Vision-Based Navigation), which is a new algorithm for vision-based navigation that has only 21 states, yet is able to track features in successive camera images and use them to propagate estimates of the spacecraft position and velocity. The filter dimension drops to 12 if attitude information is already available. The low filter dimension makes MAVeN a very reliable and practical algorithm for real-time flight implementation. The main idea is to project observed features onto a rough shape model of the ground surface, which are then used by the filter as pseudo-landmarks. The shape model is assumed to be known beforehand, as would be obtained from prior surveillance of the landing site from orbit. MAVeN does not require pre-mapped landmarks, so it is able to navigate terrain that has not been previously observed up close. This property is especially important for close proximity operations in small body missions where ground surface features are being seen for the first time at close range. MAVeN is also able to hover motionless above the ground without position error growth, which is unusual for this class of vision-based navigation algorithms. en_US
dc.description.sponsorship NASA/JPL en_US
dc.language.iso en_US en_US
dc.publisher Pasadena, CA: Jet Propulsion Laboratory, National Aeronautics and Space Administration, 2017 en_US
dc.title A Minimal State Augmentation Algorithm for Vision-Based Navigation without Using Mapped Landmarks en_US
dc.type Preprint en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record



My Account