I am Pengwei Zhang (张鹏伟), a Ph.D. student at the Institute of Automation, Chinese Academy of Sciences (CASIA). I am fortunate to be supervised by Prof. Long Cheng.
I received my Bachelor's degree in Automation from Beijing Jiaotong University. I am also a joint-training student at Zhongguancun Academy (ZGCA), working with Prof. Ce Hao.
My research focuses on building tactile-aware learning systems for reliable real-world robotic interaction, with particular interests in contact-rich manipulation, vision-tactile multimodal learning, force-aware imitation learning, and precision assembly.
Feeling the Unexpected: ResTacVLA for Contact-Rich Manipulation via Residual Tactile Representation
Pengwei Zhang, Bin Xie, Xinpan Meng, Xinyu Guo, Ce Hao, Fang Deng, Long Cheng, and Tiancai Wang
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026
Accepted
ResTacVLA reformulates touch as the residual between visual expectation and physical sensation. Predictive-coding-inspired contact primitives and a surprise-aware gate help VLA policies prioritize tactile information during visually ambiguous contact phases.
A Fine-Grained, Comprehensive, and Quantitative Rehabilitation Assessment System Based on Depth Camera
Pengwei Zhang, Long Cheng, and Yongxiang Zou
SCIENCE CHINA Technological Sciences, 2026
This work combines depth-camera motion capture with YOLOv8-Pose and an interpretable Trajectory-ROM-Smooth scoring algorithm. Evaluation with 21 stroke patients shows close agreement with experienced physicians while capturing subtle rehabilitation progress at a finer granularity.
