Autonomous navigation in deep space using optical measurements of unresolved planets and stars

Abstract

One of the essential challenges in spaceflight is spacecraft navigation. This is the estimation of the state – position, velocity, and attitude – of a spacecraft given available measurements. The development of sensors and algorithms towards this goal falls under the realm of spacecraft Guidance, Navigation, and Control (GN&C), and requires knowledge of the physics and geometry governing a specific situation. Spacecraft are generally navigated with assistance from humans on the ground, but there is increasing focus on the challenge of autonomous navigation. A spacecraft may need to be able to determine its state with no human intervention in the event of communications failure or some anomalous event, or when split-second decisions must be made in proximity of a destination several light-minutes or light-hours away from Earth. As a spacecraft travels away from Earth and other celestial bodies, into interplanetary or even interstellar space, the availability of navigation observables diminishes. This work strives to add to the body of knowledge and techniques pertaining to autonomous deep space navigation by exploring a solution to the Lost-in-Space-and-Time(LiSaT) problem using line-of-sight (LOS) measurements to unresolved planets, deriving a novel Initial Orbit Determination (IOD) algorithm using radial velocity measurements of starlight, and finding invariant properties of stars which allow for interstellar star identification and observer position estimation.