Extracting Orbital Information from the Attitude Control System of a Spacecraft near Small Bodies

Abstract

Small bodies such as asteroids and comets are becoming more and more exciting destinations. Traditional deep-space missions such as NEAR-Shoemaker or Rosetta require interaction with the ground segment on Earth for successful mission operation. Despite some a-priori information on these objects, little is known about their environment. Autonomous navigation in the vicinity of these bodies can be very challenging. Therefore, increasing the autonomy level is required to enable future deep-space missions. Several previous studies have investigated the problem of autonomous navigation for deep-space missions to small bodies. For instance, a number of simultaneous localization and mapping approaches with various sensors have been discussed and proposed. However, existing works have several assumptions or limitations. Some studies consider missions only to a particular small body or implicitly assume accurate model information. Other concepts suffer from high computational complexity, whose robust performance is not analyzed in detail. Unlike existing approaches, we investigate the feasibility of utilizing the spacecraft's attitude determination and control system (ADCS) to recover its orbital behavior. This idea has the benefit of reusing the existing equipment and algorithms. Using star trackers and gyroscope measurements, we can constrain the attitude and the angular velocity of the spacecraft relative to the celestial reference frame. Euler's equation, describing the rotational dynamics of a spacecraft, also encodes the orbital information. We measure the change in its angular momentum vector relative to the reference to infer the orbital information. While orbiting the small body, the spacecraft may conduct attitude maneuvers to leverage the determination process. The extracted orbital information can aid autonomous navigation. This paper presents the mathematical foundations of the concept and analyzes its feasibility. A number of numerical simulations are conducted in different scenarios. Finally, the robustness is assessed against disturbances and sensor noise.