端侧具身智能体研究组聚焦面向真实环境的端侧智能系统，研究可穿戴终端与机器人平台上的多模态感知、推理与控制，推动“感知—推理—行动”一体化技术在智能助手、健康监测、家庭服务与智能制造等场景落地。相关成果发表于 ACM MobiCom、MobiSys、SenSys、UbiComp、UIST 及 USENIX NSDI 等顶级会议。
面向遮挡、弱光、烟雾、多径与非视距等视觉受限场景，研究基于声学、射频与毫米波的空间感知与成像技术。通过融合物理传播机理、几何约束与场景先验，突破传统视觉感知边界，实现对空间结构与目标信息的稳定恢复，为端侧具身智能体提供更可靠的环境感知能力
2025.12.11
The human ear presents unique physiological and practical advantages for cardiac monitoring. However, existing earable systems rely on extra hardware and complex processing,making them unsuitable for resource-constrained True Wireless Stereo (TWS) earbuds. We present EarCardio, a system that ...
2025.11.21
We propose FocusX, the first mobile-deployable system achieving artifact-free all-in-focus synthesis in dynamic scenes. Our approach introduces three key innovations: 1) For focal stack acquisition, our depth prior-based dynamic focusing method that adaptively selects focus distances using real-time scene...
2025.11.21
Diffractive neural networks (NNs) have garnered attention for directly implementing wireless signal processing at the physical layer. However, they are limited by a constrained weight learning space and activation functions, which restricts their data processing capabilities. To address this, we propose an...
