Papers
2
Total Citations
10
H-Index
1
About
Shengpeng Chen is at the forefront of applying deep learning to spacecraft pose estimation, a critical technology for autonomous rendezvous, docking, and on-orbit servicing. His research bridges computer vision, control theory, and robotics to solve the extreme challenges of space environments. Chen’s most impactful work, "Spacecraft Homography Pose Estimation with Single-Stage Deep Convolutional Neural Network" (2024, 9 citations), introduces a novel single-stage CNN architecture that directly estimates homography for spacecraft pose, bypassing traditional multi-stage pipelines and achieving remarkable efficiency. Building on this, his "Quadrilateral Pose Estimation for Constrained Spacecraft Guidance and Control Using Deep Learning–Based Keypoint Filtering" (2024, 1 citation) advances the field by incorporating a keypoint filtering mechanism to enhance robustness under constrained guidance scenarios. Together, these contributions demonstrate how deep learning can overcome the limitations of classical pose estimation methods in harsh space conditions. Chen’s work is shaping the next generation of autonomous spacecraft systems, offering practical solutions for real-time, reliable pose determination that is essential for future space missions.
Research Focus
Key Achievements
Top Papers
- 1
- 2