Short-Arc Horizon-Based Optical Navigation by Total Least-Squares Estimation

Horizon-based optical navigation (OPNAV) is an attractive solution for deep space exploration missions, with strong autonomy and high accuracy. In some scenarios, especially those large variations in spacecraft distance from celestial bodies, the visible horizon arc could be very short. this case, traditional Christian–Robinson algorithm least-squares (LS) estimation inappropriate would introduce a mean residual that can even larger than standard deviation (STD). To solve problem, simplified measurement covariance model was proposed by analyzing propagation of errors. Then, unbiased element-wise total (EW-TLS) developed which equation each are fully considered. further simplify approximate generalized (AG-TLS) then proposed, achieves non-iterative using covariances. The analysis numerical simulations show algorithms have impressive advantages short-arc scenario, residuals always close to zero. Compared EW-TLS algorithm, AG-TLS trades negligible accuracy loss huge reduction execution time computing speed comparable algorithm. Furthermore, simulated scenario reveals provide reliable position estimates planetary missions.