Autonomous spacecraft timing, navigation, and control on SE(3) under uncertainty

This dissertation addresses autonomous spacecraft timing, navigation, and control under uncertainty. Future deep-space missions will require onboard systems to operate with greater autonomy under limited ground support, uncertain dynamics, and imperfect sensing. These conditions create coupled problems: timing errors affect localization, position and attitude uncertainties interact through sensing and dynamics, and control decisions must account for both the nominal trajectory and the dispersion of possible spacecraft states. This dissertation develops estimation and control methods that model these couplings directly. First, a joint orbit-clock estimation framework is derived for autonomous deep-space localization and timing. Second, a particle flow filter is developed for systems evolving on the matrix Lie group SE(3) to preserve the intrinsic geometry of coupled position-attitude states and to avoid resampling steps typical of standard particle filters. Finally, an intrinsic stochastic successive convexification method is derived for iterative covariance steering on SE(3) with chance constraints. The resulting algorithm optimizes the mean trajectory, covariance sequence, and feedback policy simultaneously. Together, these contributions advance autonomous spacecraft operation by providing uncertainty-aware methods for timing, estimation, and control in strongly coupled dynamical settings.