I. Introduction
Light pulse atom interferometers are inherently very stable. The atoms' trajectories are defined by the momentum of photons transferred to the atoms in light-matter interactions, allowing us to harness the accuracy of the wavevector for inertial measurements. Systematic effects for light-atom interactions are well understood. As a result, state-of-the-art laboratory-based atom interferometers have been consistently meeting or exceeding requirements for navigation, e.g., 10−6 and 10−4 deg/h bias stability for linear acceleration and rotation [1]. Systematic effects have been analyzed in the literature and are thus well understood, even at levels of resolution and sensitivity equal or better than required here [2], [3]. However, it remains a challenge to make such performance available in a compact package. This may require new large momentum transfer technologies, which are much less well-understood.