Robust terrain-aided localization of spacecraft using low-fidelity asteroid shape model

We proposed a localization method for spacecraft navigation in the proximity of asteroids. This method uses particle filtering with LIDAR measurement and the terrain data of an asteroid. This type of localization is called terrain-aided localization (TAL), and has been studied intensively in the fields of aeronautics and underwater robotics. However, there exist characteristic difficulties caused by the terrain uncertainty in the application of the TAL to asteroid proximity operations. To overcome these difficulties, we developed a robust terrain-aided localization method, which is effective for general TAL problems unlimited to spacecraft applications. The strength of this method lies in the capability to probabilistically address the collision of a LIDAR ray with an asteroid. The core idea of this method is to perform Monte Carlo sampling at the point at which a LIDAR ray collides with different independently simulated terrains in every trial via stochastic process. This sampling is enabled by utilizing the signed distance function. To confirm the effectiveness of the proposed method, we conducted numerical simulations. The results show the effectiveness of the proposed method.