CV

Education

Nanyang Technological University, Singapore, 2025 – present

• M.Sc. in Computer Control & Automation, School of Electrical and Electronic Engineering (EEE)
• Research in embodied AI, robot learning and dexterous manipulation

Wuhan University, Wuhan, China, 2019 – 2023

• B.S. in Optoelectronic Information Science and Engineering

Research Interests

• Embodied foundation models and VLA — large-scale multimodal manipulation policies, and RL post-training that grounds them in real interaction outcomes
• Contact-rich and force-aware manipulation — wrench-conditioned policies, sub-millimeter assembly, compliant interaction, tactile sensing
• Sample-efficient real-world reinforcement learning — human-in-the-loop learning that concentrates scarce human intervention on the most informative samples
• Agentic manipulation and embodied reasoning — structured scene representations aligning semantic planning with geometric and physical constraints, with failure diagnosis and recovery
• Dexterous teleoperation and shared autonomy — bimanual control with ego-centric perception and force feedback as an interface for contact-rich data collection

Research Experience

Embodied AI and Robot Learning, School of EEE, Nanyang Technological University, Singapore — 2025 – present

• Facet-0 — Robotic Foundation Model for Contact-Rich Precise Manipulation
  • arXiv:2609.01596, 2026. Robotic foundation model that predicts and values the contact consequences of its own actions, targeting sub-millimeter assembly where vision alone cannot distinguish a seating motion from a jamming one.
  • Joint action-wrench proposal: flow matching generates each action chunk together with the future wrist-wrench profile it is expected to induce, over a representation aligning causal wrench history with vision-language semantics and kinematic state.
  • Distributional Action-Wrench Critic: learned from deployment rollouts to separate motions with similar task progress but different contact outcomes; phase-aware rewards and contact-selective credit assignment concentrate policy improvement on decisive interactions.
  • On-robot adaptation: a lightweight bounded actor reuses the frozen representation for part-specific dynamics, with RL kept over executable Cartesian actions and an auxiliary wrench head preserving predictive, non-commanded action-contact coupling.
  • Data and results: trained on ManuFacet-1K, a 1,000-hour force-synchronized corpus spanning three embodiments and multiple manufacturing cells; 82% mean success on five sub-millimeter computer-assembly tasks versus 15% for the strongest baseline, at 0.5 mm placement accuracy and 50 ms command latency.
• UniManip — General-Purpose Zero-Shot Manipulation with Agentic Operational Graph
  • arXiv:2602.13086, 2026. Framework addressing the failure of end-to-end VLA models on long-horizon precision and of hierarchical planners on open-world semantic variation.
  • Bi-level Agentic Operational Graph: a high-level Agentic Layer for task orchestration coupled to a low-level Scene Layer for dynamic state representation, continuously aligning abstract planning with live geometric constraints.
  • Dynamic agentic loop rather than a static pipeline: object-centric scene graphs are instantiated from unstructured perception, parameterized into collision-free trajectories by a safety-aware local planner, and monitored through structured memory that autonomously diagnoses and recovers from execution failures.
  • Results: 22.5% and 25.0% higher success rate than state-of-the-art VLA and hierarchical baselines respectively; direct zero-shot transfer from fixed-base to mobile manipulation without fine-tuning or reconfiguration.
• DexTeleop-0 — Force-Aware Bimanual Dexterous Teleoperation towards Shared Autonomy
  • arXiv:2606.23431, 2026. Teleoperation framework attacking the data bottleneck in contact-rich manipulation, where embodiment gaps break kinematic mapping and tactile and force feedback are typically absent.
  • Tactile-driven adaptation strategy layered onto existing teleoperation pipelines: a real-time optimization loop translates coarse human tracking intent into precise, force-compliant robot commands.
  • Localized correction mechanism: accurate contact points are estimated and combined with a tactile-enabled fingertip force-sensing profile to compute corrections through the operational space Jacobian with respect to joint-angle updates.
  • Results: evaluated in simulation and on real hardware, consistently higher task success rates and execution efficiency than representative baselines on robust grasping, disturbance-resilient manipulation, and complex dexterous tasks.
• E2HiL — Entropy-Guided Sample Selection for Human-in-the-Loop RL (IEEE RA-L)
  • IEEE Robotics and Automation Letters, 2026; arXiv:2601.19969. Sample-efficient human-in-the-loop RL framework targeting the labor cost that limits real-world online learning.
  • Core insight: a stable, rather than merely rapid, reduction of policy entropy yields a better exploration-exploitation trade-off and higher sample efficiency.
  • Influence functions on policy entropy: each sample’s effect on entropy is modeled and efficiently estimated from the covariance of action probabilities and soft advantages of policies.
  • Moderate-influence selection: samples of moderate influence are retained while shortcut samples inducing sharp entropy drops and noisy samples with negligible effect are pruned.
  • Results: across 10 real-world manipulation tasks spanning multiple embodiments and learning frameworks, 24.9% higher success rate with 9.3% fewer human interventions than state-of-the-art HiL-RL baselines, as a policy- and embodiment-agnostic plug-and-play module.

Publications

Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation

Haoyuan Deng, Haichao Liu, Wenkai Guo, Yuan Ling, Zaijia Yang, Yuanjiang Xue, Haosheng Sun, Liangzi Wang, Ziwei Wang. (2026). "Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation." arXiv preprint arXiv:2609.01596.

DexTeleop-0: Force-Aware Bimanual Dexterous Teleoperation with Ego-Centric Perception towards Shared Autonomy

Haichao Liu, Yuyao Jiang, Hyunsun Park, Yuanjiang Xue, Ziwei Wang. (2026). "DexTeleop-0: Force-Aware Bimanual Dexterous Teleoperation with Ego-Centric Perception towards Shared Autonomy." arXiv preprint arXiv:2606.23431.

UniManip: General-Purpose Zero-Shot Robotic Manipulation with Agentic Operational Graph

Haichao Liu, Yuanjiang Xue, Yuheng Zhou, Haoyuan Deng, Yinan Liang, Lihua Xie, Ziwei Wang. (2026). "UniManip: General-Purpose Zero-Shot Robotic Manipulation with Agentic Operational Graph." arXiv preprint arXiv:2602.13086.

E2HiL: Entropy-Guided Sample Selection for Efficient Real-World Human-in-the-Loop Reinforcement Learning

Haoyuan Deng, Yudong Lin, Yuanjiang Xue, Haoyang Du, Qianzhun Wang, Boyang Zhou, Zhenyu Wu, Ziwei Wang. (2026). "E2HiL: Entropy-Guided Sample Selection for Efficient Real-World Human-in-the-Loop Reinforcement Learning." IEEE Robotics and Automation Letters.

Skills

• Programming Languages: Python, C/C++, Java
• Robot Learning: PyTorch, reinforcement learning, imitation learning, real-robot deployment
• Platforms: Linux, Windows
• Software: Matlab, Solidworks

Industry Experience

• Image Processing Engineer, TP-Link Technologies Co., Ltd., Shenzhen, China
  • Jul 2023 – Nov 2023
  • Optimized algorithms to reduce mosquito trail effects in infrared night-vision scenarios for certain high-speed PTZ camera models, enhancing image processing performance for security cameras.
  • Adjusted algorithms to reduce noise in dark areas of indoor PTZ camera models in wide dynamic range daytime scenes, mitigating noise fluctuations and improving image processing performance for security cameras.
  • Investigated clarity quantification metrics suitable for multiple scenarios.
• Hardware Development Engineer, Suzhou Sifei Technology Co., Ltd., Suzhou, China
  • Dec 2023 – Jun 2024
  • Participated in the design and development of the EVM_MSPM0L1306 learning board for the MSPM0 series M0 chip.
  • Contributed to the design and development of the EVM_MSPM0G3507 learning board for the MSPM0 series M0 chip.

Earlier Projects

Undergraduate research at Wuhan University in imaging, signal processing and embedded systems.

• Self-Developed UAV at Wuhan University, Bachelor Thesis, Wuhan University
  • Dec 2022 – May 2023
  • Completed the structural design, hardware assembly, flight control algorithm development, and flight debugging of the self-developed UAV at Wuhan University.
  • Participated in the drone gimbal debugging project with Xiangtuo Technology.
  • Assisted in the UAV video stabilization optimization work with Zhuomu Technology.
• Event Camera-Based Visual Microphone, Invention Patent, National Innovation and Entrepreneurship Project, Wuhan University
  • Mar 2021 – Mar 2023
  • Proposed an event camera-based vibration sensing method and implemented an amplitude extraction algorithm based on event point statistics. This method uses laser illumination on the object to capture laser speckle event streams, and through denoising, time segmentation, and event point counting, it efficiently reconstructs vibration signals.
  • Implemented the amplitude extraction algorithm based on event point statistics.
• Adaptive Time-Slotted Clutter Map Constant False Alarm Rate Detection Method for External Radiation Source Radar, Invention Patent, Wuhan University
  • Mar 2020 – Jul 2021
  • Developed an adaptive time division clutter map CFAR detection method for passive radar, improving false alarm performance in complex environments.
  • Divided clutter intervals based on distance-Doppler spectra, utilizing the characteristics of separated transmission and reception to optimize clutter detection in the spatial domain.
  • Established and initialized the clutter matrix, with real-time optimization through adaptive updates and forgetting factors to adapt to complex target signals and noise.
• Emotion Recognition for the Elderly Using Wearable Devices, Provincial-Level Innovation and Entrepreneurship Project, Wuhan University
  • Oct 2019 – Mar 2020
  • Designed and developed a portable emotion monitoring device, including wearable device creation, physiological data collection, and emotion recognition model training and prediction.

Competition Experience

• 5th National College Student Integrated Circuit Innovation and Entrepreneurship Competition, 3rd Prize in Central China Region, Wuhan University
  • Jul 2020 – Jul 2021
  • Modified chip IP core design, circuit design and soldering, code writing and debugging.
  • Designed wireless intelligent development board, signal and information processing, IC design, and remote wireless real-time monitoring.
• 6th National College Student Integrated Circuit Innovation and Entrepreneurship Competition, 3rd Prize in Central China Region, 1st Prize in Preliminary Round, Wuhan University
  • Jul 2021 – Jul 2022
  • Ported and accelerated target recognition based on neural networks to hardware development boards using FPGA-based CNN accelerators for SSD target detection.
• Wuhan University “Zhixing Cup” Electronic Design Competition, 1st Prize, Wuhan University
  • Jul 2020 – Sept 2020
  • Designed an intelligent tracking car with Bluetooth control, infrared tracking, ultrasonic obstacle avoidance, OLED display, and light-following docking at a designated position.

Awards and Patents

• Invention Patent: “Vibration Sensing Method and System Based on Event Camera” – Jul 2023
• Invention Patent: “Adaptive Time Division Clutter Map CFAR Detection Method for Passive Radar” – Sept 2021
• 3rd Prize in Central China Region, 5th National College Student IC Innovation and Entrepreneurship Competition – Jul 2021
• 3rd Prize in Central China Region, 6th National College Student IC Innovation and Entrepreneurship Competition – Jul 2022
• 1st Prize in Preliminary Round, 6th National College Student IC Innovation and Entrepreneurship Competition – May 2022
• 1st Prize in Wuhan University “Zhixing Cup” Electronic Design Competition – Sept 2020
• Wuhan University B-Class Scholarship – Dec 2020
• Wuhan University Excellent Student – Dec 2020
• Active Social Contributor – Jun 2020
• Wuhan University Club Cultural Festival Outstanding Individual – Sept 2022

Other Activities

• Vice President and Sanda Leader, Wuhan University Martial Arts Association – 2020 – 2021
• Member, Study Department of the SU, School of Electronics and Information, Wuhan University – 2019 – 2021
• Class Representative, Class 1906, School of Electronics and Information, Wuhan University – 2019 – 2023