Deep $\mathcal{L}^1$ Stochastic Optimal Control Policies for Planetary Soft-landing
This addresses the challenge of robust and fuel-efficient spacecraft landing, though it appears incremental by building on deep FBSDEs and stochastic search methods.
The paper tackles the Powered-Descent Guidance problem for planetary soft-landing by framing it as an L1 stochastic optimal control problem to minimize fuel consumption, and demonstrates that the controller successfully and safely lands all trajectories from a specified cone base while handling constraints and disturbances.
In this paper, we introduce a novel deep learning based solution to the Powered-Descent Guidance (PDG) problem, grounded in principles of nonlinear Stochastic Optimal Control (SOC) and Feynman-Kac theory. Our algorithm solves the PDG problem by framing it as an $\mathcal{L}^1$ SOC problem for minimum fuel consumption. Additionally, it can handle practically useful control constraints, nonlinear dynamics and enforces state constraints as soft-constraints. This is achieved by building off of recent work on deep Forward-Backward Stochastic Differential Equations (FBSDEs) and differentiable non-convex optimization neural-network layers based on stochastic search. In contrast to previous approaches, our algorithm does not require convexification of the constraints or linearization of the dynamics and is empirically shown to be robust to stochastic disturbances and the initial position of the spacecraft. After training offline, our controller can be activated once the spacecraft is within a pre-specified radius of the landing zone and at a pre-specified altitude i.e., the base of an inverted cone with the tip at the landing zone. We demonstrate empirically that our controller can successfully and safely land all trajectories initialized at the base of this cone while minimizing fuel consumption.