An Image Based Approach to Recovering the Gravitational Field of Asteroids

Andrew Melim and Frank Dellaert

In Proceedings British Machine Vision Conference 2014


Abstract NASA's DAWN spacecraft is on a mission to recover the gravity and structure of the asteroids Vesta and Ceres. Current approaches for developing a gravitational map of a celestial body rely upon use of the Deep Space Network of radio telescopes in conjunction with camera measurements for highly accurate tracking of orbiting bodies. Unfortunately, large occluding bodies greatly effect the accuracy of the radio tracking system, significantly reducing the time available for scientific experiments. This paper presents a Structure from Motion based approach that recovers the gravitational map as well as the asteroid's structure in a two step optimization process. The approach solves for a set of spherical harmonic coefficients that that define the gravitational potential given a spacecrafts relative position to the asteroid without the need for radiometric tracking from Earth based satellites. The resulting procedure can then be used to augment gravitational science during periods large noise or damaged equipment. Results are shown using the Vesta dataset from the DAWN mission and are compared with the recently published results from the DAWN gravitational team.


Poster Session


Extended Abstract (PDF, 1 page, 375K)
Paper (PDF, 11 pages, 1.0M)
Bibtex File


Andrew Melim, and Frank Dellaert. An Image Based Approach to Recovering the Gravitational Field of Asteroids. Proceedings of the British Machine Vision Conference. BMVA Press, September 2014.


	title = {An Image Based Approach to Recovering the Gravitational Field of Asteroids},
	author = {Melim, Andrew and Dellaert, Frank},
	year = {2014},
	booktitle = {Proceedings of the British Machine Vision Conference},
	publisher = {BMVA Press},
	editors = {Valstar, Michel and French, Andrew and Pridmore, Tony}
	doi = { }