Changhao Wang

University of California, Berkeley

Papers

19

Total Citations

408

H-Index

10

About

Changhao Wang is a robotics researcher whose work centers on robotic manipulation of deformable objects, legged locomotion, and robot learning. He is perhaps best known for pioneering frameworks that enable robots to reliably track, model, and control deformable linear objects (DLOs) such as cables and ropes — notoriously difficult problems due to their infinite-dimensional configuration spaces. His 2018 framework leveraging Coherent Point Drift (73 citations) laid important groundwork in this domain, followed by increasingly sophisticated approaches incorporating graph neural networks for offline-online deformation learning (62 citations) and spatial representations for cable routing (46 citations). Wang has also made notable contributions to legged locomotion, proposing online residual model learning to compensate for real-world dynamics mismatches in model-based controllers (52 citations). His more recent research expands into dual-arm manipulation in constrained 3D environments, primitive-based skill learning for robotic assembly, and SE(3)-equivariant geometric control for contact-rich tasks — reflecting a broadening toward generalizable, data-efficient robot learning. With over 370 cumulative citations across his published work, Wang has established himself as a significant voice in advancing robust, intelligent robotic manipulation systems.

Research Focus

Key Achievements

10
H-Index
19
Papers
408
Total Citations
21
Avg Citations/Paper
🏆 Most Cited Paper
A Framework for Manipulating Deformable Linear Objects by Coherent Point Drift
73 citations · 2018
📈 Most Prolific Year: 2023 (6 Papers)
🤝 Key Collaborators: 50
🏛 Institutions: University of California, Berkeley

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago