Ph.D. student at Purdue University
I am a Ph.D. candidate in the School of Aeronautics and Astronautics at Purdue University. I work with Prof. Carolin Frueh, studying techniques for monitoring space objects near the Earth for the sustainable use of space environment. My expertise lies in image processing, optical system modeling, and orbit mechanics. I also studied several computer vision algorithms in collaboration with several institutions and companies such as the Boeing Company (2023), the Lawrence Livermore National Laboratories (2022), and the Sandia National Laboratories (2021).
Ph.D. (Expected 2024), Purdue University, West Lafayette, IN
M.S. (2020), Purdue University, West Lafayette, IN
B.S. (2018), Waseda University, Tokyo, Japan
Near-Earth space environment is congested with millions of space junks. It is important to know the properties of such space junks (e.g. shape, attitude, and surface properties) to identify their sources or predict their trajectories. However, those objects are very small and they cannot be imaged by most telescopes. My research aims to recover an image of such small space junk with the help of image processing techniques.
Learn more details
I developed a software to simulate images of space objects in LEO by numerically propagting visible light from space through the free space and atmosphere based on Fourier optics. In my internship project with the Lawrence Livermore National Laboratories, I also studied the systematic distortion of optical telescopes by raytracing based on Batoid.
Learn more details
I developed or improved several computer vision algortihms:
1. Pose estimation of spacecraft based on a single image.
2. Bias correction of climate models based on domain adaptation by GAN.
3. Classification of airplanes by convolutional neural network.
Daigo Kobayashi and Carolin Frueh. “Image Recovery of LEO Objects by Leveraging Optical Turbulence and Light Curves.” Journal of Guidance, Control, and Dynamics (2024), doi: http://arc.aiaa.org/doi/abs/10.2514/1.G007634
Daniel Galea, Hsi-Yen Ma, Wen-Ying Wu, and Daigo Kobayashi. "Deep Learning Image Segmentation for Atmospheric Rivers", Artificial Intelligence for the Earth Systems 3, 1 (2024), 230048, doi: https://doi.org/10.1175/AIES-D-23-0048.1
Daigo Kobayashi, Alexander Burton, and Carolin Frueh. "AI-Assisted Near-Field Monocular Monostatic Pose Estimation of Spacecraft". In: Proceedings of the Advanced Maui Optical and Space Surveillance Technologies Conference. Sept. 2023. (PDF)
Daigo Kobayashi and Carolin Frueh. “Compressed Sensing for Enhanced Space Security: Resolving Details of Space Objects”. In: CERIAS 24th Security Symposium. Mar. 2023.
Daigo Kobayashi and Carolin Frueh. “Image-based Satellite Characterization for Low Earth Orbit”. In: 33rd AAS/AIAA Space Flight Mechanics Conference. Jan. 2023.
Daigo Kobayashi and Carolin Frueh. “Reformulating Compressed Sensing to be used with SemiResolved Point Spread Function and Light Curves for Space Object Imaging: LEO”. In: Proceedings of the Advanced Maui Optical and Space Surveillance Technologies Conference. Sept. 2022. (PDF)
Carolin Frueh, Alex Burton, Daigo Kobayashi and Liam Robinson. “Space Object Characterization from Light Curves”. In: 44th COSPAR Scientific Assembly. July 2022.
Daigo Kobayashi and Carolin Frueh. “Compressed Sensing for Satellite Characterization: Shadowing as a Sensing Matrix”. In: 8th European Conference on Space Debris. Apr. 2021. (PDF)
Daigo Kobayashi and Carolin Frueh. “Compressed Sensing for Satellite Characterization: A First Step Using Simulations”. In: 43rd COSPAR Scientific Assembly. Feb. 2021.
Daigo Kobayashi and Carolin Frueh. “Compressed Sensing for Satellite Characterization”. In: AIAA/AAS Astrodynamics Specialist Conference. Aug. 2020. (PDF)