Hao Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Hao Wang has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
Yisheng Zhong
Department of Automation Tsinghua University Beijing China
2 shared publications- Zongying Shi
Department of Automation, Tsinghua University, Beijing 100084, China
2 shared publications - Bo Feng
Apple
1 shared publication - Chunhua Shen
Zhejiang University of Technology
1 shared publication - Daniel Biggs
Peking University Peking University Peking University
1 shared publication - Feng Liu
Tsinghua University
1 shared publication
Latest Publications
- 31 July 2025
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (10) , 7460-7485
Segment Anything in Context with Vision Foundation Models
- 31 July 2025
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (10) , 7460-7485
Segment Anything in Context with Vision Foundation Models
- Open Access
Learning Deformable Body Interactions With Adaptive Spatial Tokenization
Simulating interactions between deformable bodies is vital in fields like material science, mechanical design, and robotics. While learning-based methods with Graph Neural Networks (GNNs) are effective at solving complex physical systems, they encounter scalability issues when modeling deformable body interactions. To model interactions between objects, pairwise global edges have to be created dynamically, which is computationally intensive and impractical for large-scale meshes. To overcome these challenges, drawing on insights from geometric representations, we propose an Adaptive Spatial Tokenization (AST) method for efficient representation of physical states. By dividing the simulation space into a grid of cells and mapping unstructured meshes onto this structured grid, our approach naturally groups adjacent mesh nodes. We then apply a cross-attention module to map the sparse cells into a compact, fixed-length embedding, serving as tokens for the entire physical state. Self-attention modules are employed to predict the next state over these tokens in latent space. This framework leverages the efficiency of tokenization and the expressive power of attention mechanisms to achieve accurate and scalable simulation results. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches in modeling deformable body interactions. Notably, it remains effective on large-scale simulations with meshes exceeding 100,000 nodes, where existing methods are hindered by computational limitations. Additionally, we contribute a novel large-scale dataset encompassing a wide range of deformable body interactions to support future research in this area.
Open AccessLearning Deformable Body Interactions With Adaptive Spatial Tokenization
Simulating interactions between deformable bodies is vital in fields like material science, mechanical design, and robotics. While learning-based methods with Graph Neural Networks (GNNs) are effective at solving complex physical systems, they encounter scalability issues when modeling deformable body interactions. To model interactions between objects, pairwise global edges have to be created dynamically, which is computationally intensive and impractical for large-scale meshes. To overcome these challenges, drawing on insights from geometric representations, we propose an Adaptive Spatial Tokenization (AST) method for efficient representation of physical states. By dividing the simulation space into a grid of cells and mapping unstructured meshes onto this structured grid, our approach naturally groups adjacent mesh nodes. We then apply a cross-attention module to map the sparse cells into a compact, fixed-length embedding, serving as tokens for the entire physical state. Self-attention modules are employed to predict the next state over these tokens in latent space. This framework leverages the efficiency of tokenization and the expressive power of attention mechanisms to achieve accurate and scalable simulation results. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches in modeling deformable body interactions. Notably, it remains effective on large-scale simulations with meshes exceeding 100,000 nodes, where existing methods are hindered by computational limitations. Additionally, we contribute a novel large-scale dataset encompassing a wide range of deformable body interactions to support future research in this area.
- Open Access
- 1 November 2021
- Published by Elsevier in Energy Reports
- Vol. 7, 1009-1014
Power swing detecting method using principal components analysis
- 1 November 2021
- Published by Elsevier in Energy Reports
- Vol. 7, 1009-1014
Open AccessPower swing detecting method using principal components analysis
- 8 October 2019
- Published by Wiley in Asian Journal of Control
- Vol. 23 (2) , 718-728
Robust finite‐horizon optimal control of autonomous helicopters in aggressive maneuvering
There has been an increasing need for autonomous helicopters to have the ability to perform aggressive maneuvers. To achieve this, robust and high performance controllers are required to accurately drive the helicopters. In this paper, the robust compensator (RC) technique is applied to the classic finite‐horizon LQR approach to deal with the uncertainties, including nonlinearities, external disturbances, and other interference. The robust properties of the proposed method are analyzed, and it is proved that the RC can eliminate the effects of the uncertainties as expected. The effectiveness of the proposed method is shown through simulations and real experiments of a prototype helicopter.
- 8 October 2019
- Published by Wiley in Asian Journal of Control
- Vol. 23 (2) , 718-728
Robust finite‐horizon optimal control of autonomous helicopters in aggressive maneuvering
There has been an increasing need for autonomous helicopters to have the ability to perform aggressive maneuvers. To achieve this, robust and high performance controllers are required to accurately drive the helicopters. In this paper, the robust compensator (RC) technique is applied to the classic finite‐horizon LQR approach to deal with the uncertainties, including nonlinearities, external disturbances, and other interference. The robust properties of the proposed method are analyzed, and it is proved that the RC can eliminate the effects of the uncertainties as expected. The effectiveness of the proposed method is shown through simulations and real experiments of a prototype helicopter.
- 7 March 2018
- Published by Wiley in Journal of Field Robotics
- Vol. 35 (6) , 835-849
Hierarchical fiducial marker design for pose estimation in large‐scale scenarios
In this paper, a hierarchical fiducial marker, called HArCo, is designed to guarantee a smooth pose estimation for large‐scale applications. HArCo markers have a visually identifiable structure on multiple scales, so they can be used for consistent pose estimation across a range of altitudes. Experimental results are presented to validate the performance of the proposed methodologies; oscillating platform landing experiments were conducted to show the ability of HArCo to be used in real landing tasks on the deck of a ship.
- 7 March 2018
- Published by Wiley in Journal of Field Robotics
- Vol. 35 (6) , 835-849
Hierarchical fiducial marker design for pose estimation in large‐scale scenarios
In this paper, a hierarchical fiducial marker, called HArCo, is designed to guarantee a smooth pose estimation for large‐scale applications. HArCo markers have a visually identifiable structure on multiple scales, so they can be used for consistent pose estimation across a range of altitudes. Experimental results are presented to validate the performance of the proposed methodologies; oscillating platform landing experiments were conducted to show the ability of HArCo to be used in real landing tasks on the deck of a ship.
- 1 May 2015
- Published by Elsevier in Neurocomputing
- Vol. 156, 143-150
Adaptive fuzzy control with minimal leaning parameters for electric induction motors
- 1 May 2015
- Published by Elsevier in Neurocomputing
- Vol. 156, 143-150
Adaptive fuzzy control with minimal leaning parameters for electric induction motors