Autonomous Navigation and Station-Keeping on Near-Rectilinear Halo Orbits

TL;DR

The paper develops an autonomous GNC pipeline for near-rectilinear halo orbits that integrates horizon-based optical navigation with an EKF/UKF navigation filter and two SK strategies for x-axis crossing control. It systematically compares differential correction and sequentially linearized minimization, introduces unscented-transform-based mean-state targeting, and adds hysteresis to improve robustness, all evaluated through extensive Monte-Carlo simulations. Key findings show that measurement field of view and location strongly affect navigation accuracy, that UT-based targeting and hysteresis significantly reduce cumulative SK costs, and that the end-to-end pipeline behavior exhibits a periodic covariance structure tied to NRHO geometry. The approach demonstrates viable autonomous GNC for NRHO operations and offers a pathway to extending horizon-based navigation to other libration-point orbits with minimal ground support.

Abstract

This article develops an optical navigation (OPNAV) and station-keeping pipeline for the near-rectilinear halo orbit (NRHO) in high-fidelity ephemeris model dynamics. The pipeline involves synthetic images used by the non-iterative horizon-based OPNAV algorithm, fed into an extended Kalman filter. The state estimate is used by a controller to maintain the spacecraft's motion within the vicinity of a reference NRHO. We study differential correction-based and minimization-based implementations of the x-axis crossing control scheme, and propose an improved targeting prediction scheme by incorporating the filter's state covariance with an unscented transform. We also introduce a hysteresis mechanism, which improves station-keeping cost and provides insight into the difference in performance between the differential correction-based and minimization-based approaches. We perform Monte-Carlo experiments to assess the pipeline's tracking and ΔV performances. We report several key findings, including the variability of the filter performance with the sensor field of view and measurement locations, station-keeping cost reduction achieved by the unscented transform-based prediction and hysteresis, as well as variability of the cumulative ΔV as a function of maneuver location due to the periodic structure in the OPNAV-based filter's estimation accuracy.

Figures (17)

• Figure 1: Overview of optical navigation-based station-keeping architecture and simulation environment
• Figure 2: Monte-Carlo results for measurement distributions of horizon-based OPNAV position under attitude error, using 10000 tries with $\sigma_{\rm pix} = 0.5$ pixel, $\sigma_{\phi} = 15$$\mathrm{arcsec}$. Red lines indicate empirical $2$-$\sigma$ of the samples.
• Figure 3: Range and apparent Moon diameter against osculating true anomaly along NRHO, shown for 10 superimposed revolutions
• Figure 4: NRHO shown with instantaneous true anomaly for 10 revolutions, seen from the near-side of the Moon (Earth-side observer) in the Earth-Moon rotating frame. In this view, the spacecraft tracks the orbit anti-clockwise.
• Figure 5: Measurement statistics for various camera FOV choices