Ziyan Liang
Cantonese, Chinese.

I am an undergraduate student at Sun Yat-sen University, graduating in June 2026. My undergraduate major is Electronic Information Science and Technology. From around May 2025 to the present, I have been conducting research in the SAIL laboratory under the guidance of Associate Prof. Gao Qing. From June to August 2025, I collaborated with a graduate student to complete a project, which was transformed into a paper and accepted by ICRA2026. This work mainly built a low-latency, high-precision dual-arm teleoperation system. The system mainly consists of a retargeting module and a force control module. I was primarily responsible for the code writing and overall hardware debugging of the retargeting module. From January to February 2026, under the guidance of my supervisor, I independently conducted a project to build a human-machine hand retargeting method based on spatial-temporal feature fusion. And this work was submitted to IROS2026 and is currently under review. This work primarily constructed a Transformer-based spatial-temporal encoder network to extract spatiotemporal features from human hand motion sequences. These features were then applied to dexterous hand retargeting, improving the fluency and stability of the retargeting results and reducing the dependence on input data quality. From March to April, I began writing my graduation thesis, which is an extension of my work submitted to IROS2026, incorporating reinforcement learning-based interaction strategies. From April to the present, I have been assisting a PhD student with an ongoing project on improving offline goal-conditioned reinforcement learning algorithms for multi-agent systems. My current interests focus on humanoid robot retargeting, human-demonstrated skill transfer, motion control, and interaction, particularly related to dexterous hands. I am also interested in multi-agent reinforcement learning and building multi-agent systems, especially in embodied AI area. I look forward to exchanging ideas with everyone.


Education
  • Sun Yat-sen University
    Sun Yat-sen University
    B.Eng in Electronic Information Science and Technology
    Sep. 2022 - Jul. 2026
  • Hong Kong University of Science and Technology
    Hong Kong University of Science and Technology
    M.Sc in Electronic Engineering
    Aug. 2026 - Present
News
2026
I created my first personal website!
Mar 17
Selected Publications (view all )
DexTele: A Dual-Arm Dexterous Teleoperation System Based on Motion Retargeting and Adaptive Force Control
DexTele: A Dual-Arm Dexterous Teleoperation System Based on Motion Retargeting and Adaptive Force Control

Yuanchuan. Lai, Qing Gao#, Ziyan Liang, Xianfeng Cheng, Junjie Hu, Zhaojie Ju (# corresponding author)

ICRA 2026

To address the challenges of cross-platform motion mapping and compliant interaction in dual-arm dexterous teleoperation, this work proposes DexTele, a system that achieves precise motion retargeting across heterogeneous platforms via a motion-graph encoder and latent optimization, while integrating a Vision-Language Model (VLM) with Model Predictive Control (MPC) for proactive force prediction and online optimization, demonstrating superior generalization and grasping robustness across multiple robotic hardware.

DexTele: A Dual-Arm Dexterous Teleoperation System Based on Motion Retargeting and Adaptive Force Control

Yuanchuan. Lai, Qing Gao#, Ziyan Liang, Xianfeng Cheng, Junjie Hu, Zhaojie Ju (# corresponding author)

ICRA 2026

To address the challenges of cross-platform motion mapping and compliant interaction in dual-arm dexterous teleoperation, this work proposes DexTele, a system that achieves precise motion retargeting across heterogeneous platforms via a motion-graph encoder and latent optimization, while integrating a Vision-Language Model (VLM) with Model Predictive Control (MPC) for proactive force prediction and online optimization, demonstrating superior generalization and grasping robustness across multiple robotic hardware.

All publications