Williams B.,Propulsion Laboratory |
Klein G.,Microsoft |
Reid I.,University of Oxford
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2011
Monocular SLAM has the potential to turn inexpensive cameras into powerful pose sensors for applications such as robotics and augmented reality. We present a relocalization module for such systems which solves some of the problems encountered by previous monocular SLAM systemstracking failure, map merging, and loop closure detection. This module extends recent advances in keypoint recognition to determine the camera pose relative to the landmarks within a single frame time of 33 ms. We first show how this module can be used to improve the robustness of these systems. Blur, sudden motion, and occlusion can all cause tracking to fail, leading to a corrupted map. Using the relocalization module, the system can automatically detect and recover from tracking failure while preserving map integrity. Extensive tests show that the system can then reliably generate maps for long sequences even in the presence of frequent tracking failure. We then show that the relocalization module can be used to recognize overlap in maps, i.e., when the camera has returned to a previously mapped area. Having established an overlap, we determine the relative pose of the maps using trajectory alignment so that independent maps can be merged and loop closure events can be recognized. The system combining all of these abilities is able to map larger environments and for significantly longer periods than previous systems. © 2011 IEEE.
Murchie S.,Johns Hopkins University |
Eng D.,Johns Hopkins University |
Chabot N.,Johns Hopkins University |
Guo Y.,Johns Hopkins University |
And 6 more authors.
Acta Astronautica | Year: 2014
Mars' moons Phobos and Deimos are low-albedo, D-type bodies that may preserve samples of outer solar system material that contributed organics and volatiles to the accreting terrestrial planets. A Discovery-class mission concept described in this paper, the Mars-Moon Exploration, Reconnaissance and Landed Investigation (MERLIN), will obtain in situ measurements from Deimos to test models for the moon's origin. The measurement objectives of MERLIN are to determine Deimos' elemental and miner-alogical composition, to investigate its volatile and organic content, and to characterize processes that have modified its surface. To achieve these objectives, a landed payload will provide stereo imaging and measurements of elemental and mineralogical composition and interior structure. An orbital payload will acquire global high-resolution and color imaging, putting the landing site in context by characterizing Deimos' geology. Following MOI the spacecraft flies in formation with Deimos, and uses small changes in its orbit around Mars to investigate Deimos from a range of altitudes and illuminations over 4 months. Data taken during 1- to 2-km altitude flyovers will certify a landing site. The spacecraft will be delivered to a point several kilometers above Deimos, and will navigate to landing on a fresh exposure of regolith using onboard imaging. 90 days of baseline landed operations will provide a complete set of measurements, with schedule reserve, and there is sufficient propellant to repeat the measurements at a second site. © 2013 IAA.
Gueyffier D.,ONERA |
Roux F.X.,ONERA |
Fabignon Y.,ONERA |
Chaineray G.,ONERA |
And 5 more authors.
Journal of Propulsion and Power | Year: 2015
In this paper,we present a novel numerical approach forpredicting the fluid flowin a solid rocketmotor chamberwith burning propellant grain. We use a high-order technique to track the regressing grain surface. Spectral convergence toward the exact burning surface is achieved thanks to Fourier differentiation. For the computation of the internal chamber fluid flow, we make use of a body-fitted volume mesh deforming with the grain surface. We describe several methods to deformthe volume mesh and to keep good mesh element quality without global remeshing.We then couple the surface and volume approaches and integrate them into a complex code for compressible, multispecies, turbulent flow simulations. Thanks to these methods, we are able to exhibit one of the first three-dimensional simulations of the internal flow in a realistic solid rocket motor coupled to complex grain surface regression. In prior work, burning grain surface methods have only been coupled with one-dimensional internal ballistics solvers. © 2015 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.