On Deep Recurrent Reinforcement Learning for Active Visual Tracking of Space Noncooperative Objects

TL;DR

This paper introduces RAMAVT, a deep recurrent reinforcement learning-based tracker for active visual tracking of space noncooperative objects, enhancing performance with attention mechanisms and demonstrating robustness through extensive experiments.

Contribution

The paper proposes RAMAVT, a novel deep recurrent reinforcement learning framework with attention modules for improved active visual tracking of space objects, addressing POMDP challenges.

Findings

RAMAVT outperforms existing algorithms on SNCOAT benchmark.

Incorporating MHA and SE layers improves neural network representation.

The method demonstrates high robustness and efficiency in experiments.

Abstract

Active tracking of space noncooperative object that merely relies on vision camera is greatly significant for autonomous rendezvous and debris removal. Considering its Partial Observable Markov Decision Process (POMDP) property, this paper proposes a novel tracker based on deep recurrent reinforcement learning, named as RAMAVT which drives the chasing spacecraft to follow arbitrary space noncooperative object with high-frequency and near-optimal velocity control commands. To further improve the active tracking performance, we introduce Multi-Head Attention (MHA) module and Squeeze-and-Excitation (SE) layer into RAMAVT, which remarkably improve the representative ability of neural network with almost no extra computational cost. Extensive experiments and ablation study implemented on SNCOAT benchmark show the effectiveness and robustness of our method compared with other state-of-the-art algorithm. The source codes are available on https://github.com/Dongzhou-1996/RAMAVT.