Simultaneous Localization and Mapping at Small Bodies Combining Incremental Optimization and Sequential Estimation
F. Piccolo, P. Califano, P. Panicucci, F. Topputo · Virtual Community of Pathological Anatomy (University of Castilla La Mancha) · 2025
A set of autonomous navigation algorithms for smallbody exploration is proposed, coupling pose-only factorgraph based SLAM with an EKF for velocity estimation.Monocular camera, star tracker, and LIDAR altimeter measurements are tied together in a unified framework.Rotational states are estimated using light-curves and the DAAVE algorithm, refined by a MEKF.Simulations on Bennu and camera-in-the-loop tests show estimation accuracy in the order of 10-20 m position, <0.5° attitude, and 0.1 mm/s velocity errors.Loop closures and LIDAR measurements are found to be crucial in bounding the uncertainty of the estimates.The proposed design reduces the graph complexity with respect to a fully coupled implementation and supports staged, autonomous approach operations.