Whole-Arm Manipulation

Coordinating motion and contact across the entire robot arm for human-like manipulation.

Whole-Arm Manipulation

Human manipulation is not limited to the hand. People naturally coordinate the shoulder, elbow, forearm, and hand together, using contact throughout the arm to guide motion, distribute force, and adapt to the surrounding environment.

Toward Human-Like Whole-Arm Manipulation

Building on Whole-Arm Sensing, my research focuses on turning contact awareness into coordinated action. Once the robot understands where and how its arm is interacting with the environment, it can use that information within feedback control and learning-based policies to coordinate motion and force across all of its links.

To realize this vision, my ongoing research investigates how human whole-arm behavior can be transferred to robots and gradually developed into contact-aware manipulation policies. I am currently exploring how physical interaction can serve as feedback for adapting motion and force across the entire arm, with the long-term goal of enabling robots to operate around objects, people, and constrained environments with human-like versatility and physical intelligence.

  1. Online Multi-Contact Feedback Model Predictive Control for Interactive Robotic Tasks
    • Seo Wook Han, Maged Iskandar, Jinoh Lee, Min Jun Kim
    • IEEE International Conference on Robotics and Automation (ICRA), 2024
  2. Optimization-based Multi-Contact Force Control
    • Seo Wook Han, Maged Iskandar, Jinoh Lee, Min Jun Kim
    • Korea Robotics Society Annual Conference (KRoC), 2024