Neural-Rendezvous: Provably Robust Guidance and Control to Encounter Interstellar Objects

TL;DR

Neural-Rendezvous is a deep learning framework that enables autonomous, robust, and accurate guidance and control for spacecraft to encounter fast-moving interstellar objects, with provable error bounds and real-time operation.

Contribution

The paper introduces Neural-Rendezvous, a novel deep learning-based guidance and control method with provable robustness for interstellar object encounters, outperforming traditional approaches.

Findings

Provides exponential bounds on spacecraft delivery error with high probability.

Successfully demonstrates performance on 100 ISO candidates in simulations.

Validates robustness and accuracy in UAV swarm reconfiguration scenarios.

Abstract

Interstellar objects (ISOs) are likely representatives of primitive materials invaluable in understanding exoplanetary star systems. Due to their poorly constrained orbits with generally high inclinations and relative velocities, however, exploring ISOs with conventional human-in-the-loop approaches is significantly challenging. This paper presents Neural-Rendezvous -- a deep learning-based guidance and control framework for encountering fast-moving objects, including ISOs, robustly, accurately, and autonomously in real time. It uses pointwise minimum norm tracking control on top of a guidance policy modeled by a spectrally-normalized deep neural network, where its hyperparameters are tuned with a loss function directly penalizing the MPC state trajectory tracking error. We show that Neural-Rendezvous provides a high probability exponential bound on the expected spacecraft delivery error, the proof of which leverages stochastic incremental stability analysis. In particular, it is used to construct a non-negative function with a supermartingale property, explicitly accounting for the ISO state uncertainty and the local nature of nonlinear state estimation guarantees. In numerical simulations, Neural-Rendezvous is demonstrated to satisfy the expected error bound for 100 ISO candidates. This performance is also empirically validated using our spacecraft simulator and in high-conflict and distributed UAV swarm reconfiguration with up to 20 UAVs.