Autonomous Deep Space Navigation using Variable Star Photometry and Angles-Only Navigation with a Single Camera

We present an autonomous navigation concept based on optical observations of 𝛿 Scuti
variable stars and solar system objects. Like X-ray pulsar navigation, the proposed tech-
nique uses the predicted pulsations of variable stellar sources to estimate signal time-
of-arrival, which is in turn used to estimate spacecraft position and time. In contrast to
XNAV, our method does not require specialized detectors but can instead be performed
using existing spacecraft navigation cameras or possibly star trackers. Compared to many
autonomous spacecraft navigation techniques that require prior knowledge of the space-
craft state, the proposed technique can be performed with very limited prior knowledge,
meaning that it can solve the lost-in-space-and-time problem (also known as the cold-start
problem).
Preliminary results based on simulations of the OSIRIS-APEX mission indicate that the
proposed technique can be successfully performed using existing optical instruments with
position and time accuracies on the order of 1 × 10⁻² au and 10 s. We envision this method
being used in conjunction with other higher accuracy navigation techniques that would
otherwise be infeasible in a lost-in-space-and-time scenario. In this study, we show that the
same optical instrument used to observe 𝛿 Scuti stars can also be used to perform celestial
navigation, i.e. triangulation of the spacecraft position through observations of solar
system objects, after initializing a navigation filter with position and time estimates from
our proposed variable star based technique. The complete navigation concept achieves
position and velocity 3𝜎 uncertainties on the order of 1000 km and 1 m s⁻¹.