Jonathan Wang

About

Jonathan Wang is a roboticist pushing the boundaries of how robots interact with deformable objects—specifically cables, ropes, and tethers—through dynamic, high-speed manipulation. His research centers on self-supervised learning for non-prehensile manipulation, enabling robots to perform tasks that require agility and precision beyond traditional static grasping. Wang’s major contribution is the introduction of **Planar Robot Casting (PRC)**, a novel task where a robot arm’s planar motion slides a cable’s free end toward a target, extending the robot’s effective workspace. His most-cited work, “Real2Sim2Real: Self-Supervised Learning of Physical Single-Step Dynamic Actions for Planar Robot Casting” (2022, 39 citations), demonstrates a framework that bridges simulation and reality to learn cable casting without human supervision. In his “Robots of the Lost Arc” series (2021, 38 citations), Wang shows how a UR5 robot can dynamically vault cables over obstacles, knock objects from pedestals, and weave between barriers—all learned autonomously. With over 100 total citations, his work has immediate applications in cable management, search-and-rescue, and industrial automation, establishing him as a leading voice in dynamic, self-supervised robotic manipulation.

Research Focus

Key Achievements

Top Papers

1. 1
   Real2Sim2Real: Self-Supervised Learning of Physical Single-Step Dynamic Actions for Planar Robot Casting

   39 citations · 2022
2. 2
   Robots of the Lost Arc: Self-Supervised Learning to Dynamically Manipulate Fixed-Endpoint Cables

   38 citations · 2021
3. 3
   Planar Robot Casting with Real2Sim2Real Self-Supervised Learning

   13 citations · 2021
4. 4
   Robots of the Lost Arc: Learning to Dynamically Manipulate Fixed-Endpoint Ropes and Cables.

   6 citations · 2020
5. 5
   Robots of the Lost Arc: Self-Supervised Learning to Dynamically Manipulate Fixed-Endpoint Cables

   2 citations · 2020

Key Collaborators

Contact & Links