Notice Site under construction
Work Publications About Contact
Currently · Italy
PhD Student Italian Institute of Technology, Italy 2025 —
DOWON PARK

made with CLIPasso

Soft Robotics. Human-robot coexistence.
Robotics at the boundary of culture & art.

Explore work → GitHub ↗

Research

Research
activities
SoftHand IIT/Pisa Work in progress
Ongoing
POINT CLOUD unknown bottle BASIC SHAPES bounding boxes GRASPING POSE QUALITY rank & select EXECUTE selected grasp
From an unknown object's point cloud to the selected grasp — GLP 2.0 pipeline.

Approach

Given the point cloud of an unknown object, GLP 2.0 decomposes it into basic shapes and uses a model learned from human demonstrations (6-DoF poses + interaction forces) to generate candidate grasps. A novel analytical grasp quality score then ranks them, accounting for the gripper used, collisions with the environment, and estimated interaction forces.

Results

  • 94.0% success rate — 150 grasps on 30 objects with the Pisa/IIT SoftHand
  • 85.0% success rate — 80 grasps on 16 objects with the Franka Emika Hand
  • +25% improvement over the previous version (GLP 1.0)
  • Outperforms two state-of-the-art baselines on both grippers
  • ICRA 2024 — IEEE T-RO King-Sun Fu Best Paper Award, Honorable Mention
Read paper ↗ Watch on YouTube ↗

Publications

Publications
View all on Google Scholar ↗

About

Do Won Park
Soft Robotics Grasping HRI ART Cultural Heritage TELEOPERATION

Do Won Park received the B.Sc. in Bioengineering and the M.Sc. in Robotics and Automation Engineering from the University of Pisa, Italy, in 2017 and 2021. His master's thesis investigated grasp planning algorithms for robots equipped with anthropomorphic hands.

Following his studies, he spent three years in industrial automation at a multinational company before returning to research. He subsequently joined the Research Center "E. Piaggio" at the University of Pisa as a Research Fellow. Since 2025, he is a PhD Student at the Italian Institute of Technology (IIT), Genova, within the Soft Robotics for Human Cooperation and Rehabilitation unit.

His research spans soft robotic grasping, human-robot interaction, and the application of robotics to art and cultural heritage, investigating the capabilities of machines within unconventional domains.

He contributed to RePAIR (EU H2020) and is currently involved in the robotic component of AUTOMATA (EU Horizon Europe, 2024–2029).

🏅
Honorable Mention — ICRA 2024, Yokohama IEEE T-RO King-Sun Fu Memorial Best Paper Award, as co-author of a paper on robotic grasping with anthropomorphic hands.
Grasp It Like a Pro 2.0.
ICRA 2024, Yokohama — King-Sun Fu Best Paper Award, Honorable Mention ceremony
GET
IN TOUCH

Open to collaborations, conversations, and unconventional ideas.