ACT-SLAM: Active Continuous-Time SLAM for Powered Descent Guidance Maneuvers

Illustration demonstrating the modeling approach behind the ACT-SLAM technique applied to the rocket landing (powered descent guidance) problem.

Abstract

The problem of guidance-navigation co-design for autonomous aerospace systems concerns the simultaneous satisfaction of guidance and navigation requirements in mission & trajectory design, and is of keen interest for the realization of robust autonomy in the aerospace industry. In this paper, a general-purpose technique–Active Continuous-Time Simultaneous Localization & Mapping (ACT-SLAM)–is proposed to systematically and flexibly solve for various elements of guidance-navigation co-design and is specialized for scenarios involving spatial perception objectives. The resulting architecture makes use of advances in successive convexification for nonconvex trajectory optimization to solve the resulting problem at near-real-time speeds. ACT-SLAM is consequently demonstrated on the highly-constrained powered descent guidance (PDG) problem in a lunar environment with LiDAR-based measurements, showing promising results for joint reduction in vehicle and mapping uncertainty. Furthermore, rigorous satisfaction of both guidance and navigation requirements is shown through 3-sigma Monte Carlo analysis, and benchmark comparisons to several recent methods are made, demonstrating considerable improvements in terms of information-theoretic objectives.(Buckner et al., 2026)

A simulation of a landing maneuver generated by ACT-SLAM. On the left, the onboard LiDAR sensor's field-of-view is depicted by a cone emitting from the lander, color coded by ground landmarks as they become visible. On the right, elements of the state estimator's covariance matrix (which measures uncertainty of both the lander's position and the landmarks' positions) are color-coded according to relative magnitude, with darker elements corresponding to higher uncertainty. As landmarks come in to field of view, elements on the diagonal of the matrix become brighter indicating an improved navigational solution, which is quantified by the information gain, ΔI(t), relative to the beginning of the maneuver.

A comparison between the passive perception baseline, info-PDG and ACT-SLAM for the same landing scenario; ACT-SLAM outperforms the compared methods by between 29% and 36% in terms of information gain.

Photo of me winning the best paper award (GNC Student Paper Competition) at SciTech.

Abstract

References

2026