Estimation of asteroid model parameters using particle filters

Planning of open-loop maneuvers in an asteroid's dynamic environment requires the ability to estimate the forces acting on the spacecraft as a function of position and time. These forces can be estimated using a mathematical model of the asteroid provided that the model parameters may be accurately inferred. In this paper we demonstrate how a particle filter can quickly and accurately estimate these model parameters using the spacecraft's average commanded thrust whilst hovering at two locations in close proximity to the asteroid. The accuracy of the particle filter is then compared to that of two state of the art optimization algorithms. We also show how this method can easily be extended to estimate internal gradients in the asteroid's density.