Jeon Ho Kang
Jeon Ho Kang

Ph.D. Student in Robotics

University of Southern California

I am a Ph.D. student in Robotics at the University of Southern California, fortunate to be advised by Professor S.K. Gupta. I plan to graduate in December 2026. I am a member of the RROS Lab, where we focus on robot learning for contact-rich manipulation.

My work also spans VLA post-training at Amazon Fauna and real2sim2real at Amazon Robotics for general robotic intelligence. I am particularly interested in developing multimodal learning frameworks that integrate force, tactile inputs, and language.

I have authored three first-author publications, including papers in IEEE ICRA, RA-L, and CASE. Additionally, I have contributed to three ASME conference papers (MSEC and IDETC), with our MSEC 2024 paper receiving the Best Conference Paper Award (2nd place). I have also served as a reviewer for conferences and journals including ICRA, RA-L, IROS, CASE, and Humanoids.

Download CV
Interests
  • Deep Learning
  • Robotics
  • Sensory Fusion
  • Imitation Learning
  • Motion Planning
  • Task Planning
Education

  • Ph.D. in AME (Robotics) 2026

    University of Southern California

  • MS in Mechanical Engineering (Automation) 2023

    University of Southern California

  • BS Mechanical Engineering 2022

    University of Southern California

News

  • September 2026: Our paper titled Vision-Language Grounded Task-Context-Aware Imitation Learning for Robotic Disassembly has been accepted to IEEE Robotics and Automation Letters (RA-L)!
  • May 2026: Joined Amazon as an Applied Scientist II Intern!
  • April 2025: Now our paper, code and dataset can be accessed via Arxiv, Papers with code, and Hugging Face !
  • March 2025: Looking forward to interning at Honda Resarch Institute this summer as Resesarch Scientist Intern working on Behavior Models for Dexterous Manipulation!
  • March 2025: Our paper on Compliant Object prying has been accepted on Robotics and Automation Letters!
  • March 2024: Our paper at MSEC 2024 won the Best Conference Paper Award!
  • February 2024: Our paper on Lage Language Moddels for Contingency Recovery and Task Allocation has been accepted for publication at ICRA 2024!
Research

Experience

  1. Amazon Robotics - Vulcan Stow Behaviors Team logo

    Applied Scientist II Intern

    Amazon Robotics - Vulcan Stow Behaviors Team

    Aug 2026 – Present

    Contact-rich insertion and sweeping using real2sim2real RL

  2. Amazon - Fauna Team (Humanoids) logo

    Applied Scientist II Intern

    Amazon - Fauna Team (Humanoids)

    May 2026 – Aug 2026

    VLA post-training using reinforcement learning (RL).

  3. Honda Research Institute logo

    Research Scientist Intern

    Honda Research Institute

    May 2025 – Aug 2025

    Researh on vision language action models applied on dexterous manipulations tasks

  4. Center for Advanced Manufacturing logo

    Graduate Researcher

    Center for Advanced Manufacturing

    Jan 2022 – Present

    Conduct research on decision-making for robotic manipulation and task planning in manufacturing, and publish findings in top-tier journals and conferences.

  5. Engineering Intern (Undergraduate)

    Versa Products

    Jan 2022 – May 2022

Publications
Jeon Ho Kang, Igal Tamarkin, Ethan Niu, Ian Novales, Satyandra K. Gupta (2026). Vision-Language Grounded Task-Context-Aware Imitation Learning for Robotic Disassembly. IEEE Robotics and Automation Letters, 2026 (accepted for publication).
PDF Cite Project DOI
Jeon Ho Kang, Sagar Joshi, Neel Dhanaraj, Satyandra K. Gupta (2025). Task-Context-Aware Diffusion Policy with Language Guidance for Multi-task Disassembly. 2025 IEEE 20th International Conference on Automation Science and Engineering (CASE).
PDF Cite Code Project Video
Jeon Ho Kang, Sagar Joshi, Ruopeng Huang, Satyandra K. Gupta (2025). Robotic Compliant Object Prying Using Diffusion Policy Guided by Vision and Force Observations. Robotics and Automation Letters, 2025.
PDF Cite Code Dataset Project Video
Rishabh Shukla, Samrudh Mude, Raj Talan, Neel Dhanaraj, Jeon Ho Kang, Satyandra K. Gupta (2025). Force-Conditioned Diffusion Policies for Compliant Sheet Separation Tasks in Bimanual Robotic Cells. ICRA 2025.
Cite Video
Jeon Ho Kang, Neel Dhanaraj, Siddhant Wadaskar, Satyandra K. Gupta (2024). Generating and Applying Contingency Handling Procedures in Human-Robot Teams in Manufacturing Applications. IEEE International Conference on Robotics and Automation (ICRA), 2024.
PDF Cite Video
Jeon Ho Kang, Neel Dhanaraj, Omey Manyar, Siddhant Wadaskar, Satyandra K. Gupta (2024). A Task Allocation and Scheduling Framework to Facilitate Efficient Human-Robot Collaboration in High-Mix Assembly Applications. [Best Paper Award] Proceedings of ASME’s Manufacturing Science and Engineering Conference MSEC 2024 June 17-June 21, 2024, Knoxville TN, USA.
PDF Cite
Jeon Ho Kang, Rishabh Shkla, Moksh Mehta, Satyandra K. Gupta (2024). A Learning Framework for Enabling Robots to Autonomously Dispense Granular Material On-Demand . DETC 2024 August 25–28, 2024 Washington, DC, USA.
PDF Cite
Neel Dhanaraj, Jeon Ho Kang, Anirban Mukherjee, Heramb Nemlekar, Stefanos Nikolaidis, Satyandra K. Gupta (2024). Multi-robot task allocation under uncertainty via hindsight optimization. 2024 IEEE International Conference on Robotics and Automation (ICRA).
PDF Cite
Neel Dhanaraj, Minseok Jeon, Jeon Ho Kang, Stefanos Nikolaidis, Satyandra K. Gupta (2024). Preference Elicitation and Incorporation for Human-Robot Task Scheduling. 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE).
Cite
Jeon Ho Kang, Paolo Limcaoco, Neel Dhanaraj, Satyandra K. Gupta (2023). Safe Robot to Human Tool Handover to Support Effective Collaboration. In.
PDF Cite Video
Projects
Skills & courses
Technical Skills
Programming

C/C++, Python, Linux

Libraries

Pytorch, TensorFlow, MoveIt, ROS/ROS2,
OpenCV, Open3D

Simulation Software

Isaac Sim, Mujoco, Webots, Gazebo

Robots

Kuka, ABB, Yasakawa, UR

Developer Tools

Git, CUDA, Docker

Courses
CS 182 - Designing, Visualizing and Understanding
Deep Neural Networks

Great course taught by Professor Sergey Levine: Syllabus. Spans from basic affine transform, regularization techniques. Then goes in to different architectures like CNN, RNN and Transformers. Touches on different applications like natural language processing, computer vision and imitation learning for self-driving cars and robotics.

NN-zero-to-hero

Considered one of the best tutorials to neural networks taught by Andrej Karpathy.

AME 540 - Probability and Statistics for Engineering Science

Course taught at USC by Professor Assad Obeai. Touches on introductory theories behind probability like random variables and vectors, Conditional distributions and Bayes theorem, and stocahstic processes and its applications

Accomplishments
Best Conference Paper Award (2nd Place)
The American Society of Mechanical Engineers - ASME ∙ November 2023
NSF Travel Award 2024
National Science Foundation (NSF) ∙ September 2023
Research Featured in IEEE Video Friday
IEEE Spectrum ∙ March 2023