Deep ℒ^1 Stochastic Optimal Control Policies for Planetary Soft-landing

09/01/2021
by   Marcus A. Pereira, et al.
6

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 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.

READ FULL TEXT

page 15

page 17

page 19

research
09/02/2020

Safe Optimal Control Using Stochastic Barrier Functions and Deep Forward-Backward SDEs

This paper introduces a new formulation for stochastic optimal control a...
research
02/11/2019

Neural Network Architectures for Stochastic Control using the Nonlinear Feynman-Kac Lemma

In this paper we propose a new methodology for decision-making under unc...
research
04/12/2022

A deep learning method for solving stochastic optimal control problems driven by fully-coupled FBSDEs

In this paper, we mainly focus on the numerical solution of high-dimensi...
research
06/11/2019

Deep 2FBSDEs for Systems with Control Multiplicative Noise

We present a deep recurrent neural network architecture to solve a class...
research
06/11/2020

Stochastic properties of an inverted pendulum on a wheel on a soft surface

We study dynamics of the inverted pendulum on the wheel on a soft surfac...
research
06/22/2020

Non-convex Optimization via Adaptive Stochastic Search for End-to-End Learning and Control

In this work we propose the use of adaptive stochastic search as a build...
research
10/02/2020

Memory Clustering using Persistent Homology for Multimodality- and Discontinuity-Sensitive Learning of Optimal Control Warm-starts

Shooting methods are an efficient approach to solving nonlinear optimal ...

Please sign up or login with your details

Forgot password? Click here to reset