Deep $\mathcal{L}^1$ Stochastic Optimal Control Policies for Planetary Soft-landing

TL;DR

This paper presents a deep learning-based stochastic optimal control method for planetary soft-landing, capable of handling complex constraints and nonlinear dynamics without linearization, demonstrating robust and fuel-efficient landings.

Contribution

It introduces a novel deep FBSDE-based approach for stochastic optimal control in planetary landing, avoiding convexification and linearization of constraints and dynamics.

Findings

Successfully lands all trajectories within the cone safely.

Minimizes fuel consumption effectively.

Robust to stochastic disturbances and initial conditions.

Abstract

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.