NeRF-based Simultaneous Pose Estimation and 3D Reconstruction for Non-Cooperative Space Target

Abstract

Accurate pose estimation and 3D reconstruction of Non-Cooperative Space Targets (NCSTs) are critical for proximity operations in active debris removal and on-orbit servicing. In this paper, we propose a novel NeRF-based Simultaneous Pose Estimation and 3D Reconstruction (SPAR) framework to address the challenges of efficiency and reliability in traditional point-based methods, which contains three key components, \ie a multi-resolution hash encoder to reduce computational cost, a 2D keyframe feature enhancement to guide view generalization, and a direct photometric constraint to stabilize pose estimation. Furthermore, the proposed framework is evaluated on a newly constructed Spacecraft Pose Estimation and 3D Reconstruction Dataset (SPARD), comprising both synthetic and real RGB-D images and the experimental results demonstrate its effectiveness. Our framework achieves real-time processing at 51 Hz with a pose estimation accuracy of 1.26 cm translation error and 0.97° rotation error. The results show improvements over traditional and NeRF-based baselines, validating its applicability to space missions with NCST.

Pipeline

Spacecraft Pose Estimation and 3D Reconstruction Dataset
(SPARD)

Synthetic Scene Setup

Real Scene Setup

Experiment Results

Pose Estimation

(a) The circular trajectory

(b) The conical spiral trajectory

3D Reconstruction

(a) Smart Satellite 1A (Real)

(b) Smart Satellite 1A (Virtual)

(c) Venus Express

(d) Proba-2

BibTeX

@article{Yang2025NSPAR,
  author    = {Dazhuang, Yang and Yizhai, Zhang and Ge, Yu and Jiafu, Jiao and Binglu, Wang and Panfeng, Huang},
  title     = {NeRF-based Simultaneous Pose Estimation and 3D Reconstruction for Non-Cooperative Space Target},
  year      = {2025},
}