Chen Chen

Chen Chen 陈晨

Senior Research Associate · Hangzhou Innovation Institute, Beihang University 高级副研究员 · 北京航空航天大学杭州创新研究院

14publications论文
280+citations总引用
7granted patents授权发明专利

I work on embodied AI and robotics — semantic object-goal navigation, SLAM and 3D reconstruction in dynamic environments, and LLM-powered robot systems. I received my M.Eng. in Computer Science from Beihang University (2022), advised by Prof. Jianwei Niu and Assoc. Prof. Xuefeng Liu, with a Master's thesis on medical image analysis (first-author paper in IEEE TMI), and my B.Eng. from Taiyuan University of Technology. 我的研究方向是具身智能与机器人:语义目标导航、动态环境下的 SLAM 与三维重建、大模型驱动的机器人系统。2022 年于北京航空航天大学获硕士学位(保研,导师:牛建伟教授刘雪峰副教授),硕士课题为医学图像分析(一作论文发表于 IEEE TMI);本科毕业于太原理工大学计算机科学与技术专业。

What I work on研究方向

Embodied AI具身智能

Semantic Navigation语义目标导航

Zero-shot object-goal navigation with vision-language models: adaptive multi-cue frontier maps (KSEM'26), commonsense & memory-augmented object search (ICRA'26), 3D scene graph maintenance during task execution (ICRA'25), and LLM-enhanced navigation (L2R-Nav, KSEM'24). 2nd inventor on a granted patent for multi-cue semantic-matching goal navigation. 基于视觉-语言模型的零样本物体目标导航:多维线索自适应加权前沿地图(KSEM'26)、常识与记忆增强的物体搜索(ICRA'26)、任务执行中的三维场景图维护(ICRA'25)、LLM 增强导航框架(L2R-Nav,KSEM'24)。「多线索语义匹配的目标导航」授权发明专利第二发明人。

3D Vision三维视觉

SLAM in Dynamic Scenes动态场景 SLAM 与重建

Making neural implicit and 3D Gaussian Splatting SLAM robust to moving people and objects: NID-SLAM (ICME'24), GLC-SLAM, DL-SLAM. Ongoing work on dynamic-object handling for feed-forward 3D reconstruction. 让神经隐式与 3D Gaussian Splatting SLAM 在有人员走动的动态环境中保持鲁棒:NID-SLAM(ICME'24)、GLC-SLAM、DL-SLAM。在研:面向前馈式三维重建的动态物体处理。

Robotics机器人系统

Robot Systems Engineering机器人系统工程

Built the perception–mapping–navigation stack of a self-developed composite robot (differential base + SCARA arm) on RK3588 within a ¥12M Zhejiang Key R&D project: EKF multi-sensor fusion, 2D-laser & RTAB-Map SLAM, PID fine docking (<2cm/<3°), vision-based grasp-pose generation, LLM task decomposition (>95% grasp success), and a self-developed web robot visualization console (rviz-like). Deployed in automated laboratories and outdoor inspection. Currently fine-tuning and deploying VLA policies (π0-family) on a self-developed dual-arm robot with edge deployment on NVIDIA AGX Orin/Thor. 在 1200 万浙江省重点研发项目中负责自研复合机器人(差速底盘 + SCARA 臂)的感知-建图-导航全栈(RK3588):EKF 多传感器融合、2D 激光与 RTAB-Map SLAM、PID 精定位(<2cm/<3°)、基于视觉的抓取位姿生成、LLM 任务拆解抓取(>95%)、自研 Web 机器人可视化(类 rviz)。已落地自动化实验室与室外巡检。当前:在自研双臂机器人上微调与部署 π0 系 VLA 策略,并在 NVIDIA AGX Orin/Thor 上进行端侧部署。

Medical AI医疗影像

Medical Image Analysis医学图像分析

Earlier line of work (2019–2023): multi-modal breast ultrasound analysis — domain-knowledge-guided CEUS video diagnosis (IEEE TMI 2021, first author, 150+ citations) and US+CEUS mutual-aid lesion segmentation (BIBM 2023, co-first author), with two granted patents on ultrasound classification and lesion detection. 早期方向(2019–2023):多模态乳腺超声分析——融合放射科医生领域知识的 CEUS 视频良恶性诊断(IEEE TMI 2021 一作,引用 150+)、US+CEUS 互助病灶分割(BIBM 2023 共同一作);配套超声图像分类与病灶检测授权发明专利 2 项。

Publications论文

Also on Google Scholar. * = co-first author, † = corresponding author. 完整列表见 Google Scholar。* 为共同一作,† 为通讯作者。

From Glance to Inspection: Frontier Maps from Adaptive Weighting of Multi-Dimensional Cues for Zero-Shot Object Navigation

Qingfeng Li, Chen Chen, Xiaoze Wu, Xiaozheng Xie, Ningbo Gu, Jianwei Niu

KSEM · 2026

to appear已录用

CMAR-search: Commonsense and Memory Augmented Reasoning for Object Search in Dynamic Interactive Environments

Kaiyao Liao, Qingfeng Li, Xinlei Zhang, Chen Chen, Qing Sun, Jianwei Niu

ICRA · 2026

to appear已录用

SV-Plan: En-Route Task Planning Using Semantic Voronoi Graph

Mingxi Wang, Qingfeng Li, Chen Chen, Ningbo Gu, Xinlei Zhang, Kaiyao Liao, Mu Li

KSEM · 2026

to appear已录用

SAC-SLAM: Building 3D Spatial Knowledge Bases via Surface-Aligned Consistent Mapping

Qi He, Qingfeng Li, Ziheng Xu, Chen Chen, Ningbo Gu, Feng Hong

KSEM · 2026

to appear已录用

DL-SLAM: Enabling High-Fidelity Gaussian Splatting SLAM in Dynamic Environments via Dual-Level Probability

Ziheng Xu, Qingfeng Li, Xuefeng Liu, Chen Chen, Jianwei Niu

arXiv 2607.01860 · 2026

Interaction-Driven Updates: 3D Scene Graph Maintenance During Robot Task Execution

Qingfeng Li, Xinlei Zhang, Chen Chen†, Haochen Zhao, Jianwei Niu

ICRA · 2025

Corresponding author通讯作者DOI
NID-SLAM

NID-SLAM: Neural Implicit Representation-based RGB-D SLAM in Dynamic Environments

Ziheng Xu, Jianwei Niu, Qingfeng Li, Tao Ren, Chen Chen

ICME · 2024

40+ citations引用 40+arXiv
L2R-Nav

L2R-Nav: A Large Language Model-Enhanced Framework for Robotic Navigation

Xiaoze Wu, Qingfeng Li, Chen Chen, Xinlei Zhang, Haochen Zhao, Jianwei Niu

KSEM · 2024

GLC-SLAM: Gaussian Splatting SLAM with Efficient Loop Closure

Ziheng Xu, Qingfeng Li, Chen Chen, Xuefeng Liu, Jianwei Niu

arXiv 2409.10982 · 2024

IMAN

IMAN: An Iterative Mutual-Aid Network for Breast Lesion Segmentation on Multi-modal Ultrasound Images

Xiaozheng Xie, Chen Chen*, Xuefeng Liu, Yong Wang, Rui Wang, Jianwei Niu

BIBM · 2023

Co-first author共同一作

DGANet: A Dual Global Attention Neural Network for Breast Lesion Detection in Ultrasound Images

Hui Meng, et al., Chen Chen

Ultrasound in Medicine & Biology · 2023

30+ citations引用 30+

Conventional Ultrasound and Contrast-Enhanced Ultrasound Radiomics in Breast Cancer and Molecular Subtype Diagnosis

Xuantong Gong, et al., Chen Chen

Frontiers in Oncology · 2023

30+ citations引用 30+

Anatomical Landmarks Annotation on 2D Lateral Cephalograms with Channel Attention

Dongfeng Du, Tao Ren, Chen Chen, Yiran Jiang, Guangying Song, Qingfeng Li, Jianwei Niu

IEEE/ACM CCGRID · 2022

TMI

Domain Knowledge Powered Deep Learning for Breast Cancer Diagnosis Based on Contrast-Enhanced Ultrasound Videos

Chen Chen, Yong Wang, Jianwei Niu, Xuefeng Liu, Qingfeng Li, Xuantong Gong

IEEE Transactions on Medical Imaging (TMI) · 2021 · 40(9): 2439–2451

First author一作 150+ citations引用 150+ IEEE

Granted Invention Patents (CN)授权发明专利

Experience & Education工作与教育经历

Open Source开源贡献

Selected Awards主要奖项