Yinan Deng

Yinan Deng

Auto-generated profile, based on publicly available information

Yinan Deng

School of Automation, Beijing Institute of Technology, Beijing, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 August 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YDYinan Deng
Yinan Deng
JWJiahui Wang
Jiahui Wang
JZJingyu Zhao
Jingyu Zhao
XTXinyu Tian
Xinyu Tian
GCGuangyan Chen
Guangyan Chen
YYYufeng Yue
Yufeng Yue
Environment representations endowed with sophisticated semantics are pivotal for facilitating seamless interaction between robots and humans, enabling them to effectively carry out various tasks. Open-vocabulary representation, powered by Visual-Language models (VLMs), possesses inherent advantages, including zero-shot learning and open-set cognition. However, existing open-vocabulary maps are primarily designed for small-scale environments, such as desktops or rooms, and are typically geared towards limited-area tasks involving robotic indoor navigation or in-place manipulation. They face challenges in direct generalization to outdoor environments characterized by numerous objects and complex tasks, owing to limitations in both understanding level and map structure. In this work, we propose OpenGraph, a novel open-vocabulary hierarchical graph representation designed for large-scale outdoor environments. OpenGraph initially extracts instances and their captions from visual images, enhancing textual reasoning by encoding captions. Subsequently, it achieves 3D incremental object-centric mapping with feature embedding by projecting images onto LiDAR point clouds. Finally, the environment is segmented based on lane graph connectivity to construct a hierarchical representation. Validation results from SemanticKITTI and real-world scene demonstrate that OpenGraph achieves high segmentation and query accuracy.

20 March 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YDYinan Deng
Yinan Deng
YTYujie Tang
Yujie Tang
YYYi Yang
Yi Yang
DWDanwei Wang
Danwei Wang
YYYufeng Yue
Yufeng Yue
Collaborative mapping aids agents in achieving an efficient and comprehensive understanding of their environment. Recently, there has been growing interest in using neural networks as maps to represent functions that implicitly define the geometric features of a scene. However, existing implicit mapping algorithms are constrained to single-agent scenarios, thus restricting mapping efficiency. In this paper, we present MACIM , a M ulti- A gent C ollaborative I mplicit M apping algorithm to construct implicit Euclidean Signed Distance Field (ESDF) maps, formulated as a distributed optimization task. In our formulation, each agent independently maintains its own local data and neural network. At each iteration, agents train networks using local data and network weights from their peers. Subsequently, they transmit the latest version of network weights to their neighbors, thus keeping the local network weights of all agents continuously consistent. When optimizing the network model, the agents use not raw but in-grid fused sensor data to prevent training data conflicts. In addition, we constrain the signed distance values of unobserved regions by Small Batch Euclidean Distance Transform (SBEDT) to mitigate reconstruction artifacts. The verification results on multiple scenes demonstrate that MACIM builds more accurate ESDFs and meshes than single-agent strategy and some distributed optimization methods. The source code of MACIM is publicly available at https://github.com/BIT-DYN/MACIM.

5 April 2023 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Open access
YDYinan Deng
Yinan Deng
MWMeiling Wang
Meiling Wang
YYYi Yang
Yi Yang
DWDanwei Wang
Danwei Wang
YYYufeng Yue
Yufeng Yue
Generating an accurate and continuous semantic occupancy map is a key component of autonomous robotics. Most existing continuous semantic occupancy mapping methods neglect the potential differences between voxels, which reconstruct an overinflated map. What is more, these methods have high computational complexity due to the fixed and large query range. To address the challenges of overinflation and inefficiency, this article proposes a novel sharp-edged and efficient continuous semantic occupancy mapping algorithm (SEE-CSOM). The main contribution of this work is to design the Redundant Voxel Filter Model (RVFM) and the Adaptive Kernel Length Model (AKLM) to improve the performance of the map. RVFM applies context entropy to filter out the redundant voxels with a low degree of confidence, so that the representation of objects will have accurate boundaries with sharp edges. AKLM adaptively adjusts the kernel length with class entropy, which reduces the amount of data used for training. Then, the multientropy kernel inference function is formulated to integrate the two models to generate the continuous semantic occupancy map. The algorithm has been verified on indoor and outdoor public datasets and implemented on a real robot platform, validating the significant improvement in accuracy and efficiency.

23 July 2025 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YTYujie Tang
Yujie Tang
MWMeiling Wang
Meiling Wang
YDYinan Deng
Yinan Deng
ZZZibo Zheng
Zibo Zheng
JDJingchuan Deng
Jingchuan Deng
YYYufeng Yue
Yufeng Yue
In daily domestic settings, frequently used objects like cups often have unfixed positions and multiple instances within the same category, and their carriers also frequently change. As a result, it becomes challenging for a robot to efficiently navigate to a specific instance. To tackle this challenge, the robot must capture and update scene changes and plans continuously. However, current object-navigation approaches primarily focus on the semantic level and lack the ability to dynamically update the scene representation. In contrast, this paper captures the relationships between frequently used objects and their static carriers. It constructs an open-vocabulary Carrier-Relationship Scene Graph (CRSG) and updates the carrying status during robot navigation to reflect the dynamic changes of the scene. Based on CRSG, we further propose an instance navigation strategy that models the navigation process as a Markov Decision Process. At each step, decisions are informed by the Large Language Model's commonsense knowledge and visual-language feature similarity. We designed a series of long-horizon navigation tasks for frequently used everyday items in the Habitat simulator. The results demonstrate that by updating the CRSG, the robot can navigate efficiently to moved targets. Additionally, we conducted extensive experiments on a real robot, demonstrating the effectiveness of our method and exploring its limitations. The project page can be found here: https://OpenIN-nav.github.io.

4 December 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YDYinan Deng
Yinan Deng
JWJiahui Wang
Jiahui Wang
JZJingyu Zhao
Jingyu Zhao
JDJianyu Dou
Jianyu Dou
YYYi Yang
Yi Yang
YYYufeng Yue
Yufeng Yue
In recent years, there has been a surge of interest in open-vocabulary 3D scene reconstruction facilitated by visual language models (VLMs), which showcase remarkable capabilities in open-set retrieval tasks. Although the semantic ambiguity of existing point-wise feature maps is alleviated by open-vocabulary mask segmenters for object-level understanding, effectively retaining fine-grained features within objects simultaneously remains challenging. To address these challenges, we introduce OpenObj, an innovative approach to build open-vocabulary object-level Neural Radiance Fields (NeRF) with fine-grained understanding. In essence, OpenObj establishes a robust framework for efficient and watertight scene modeling and comprehension at the object level. Specifically, we obtain cross-frame consistent instance-level masks for supervision through our two-stage mask clustering module. Moreover, by incorporating part-level features into the object NeRF models, OpenObj not only captures object-level instances but also preserves an understanding of their internal granularity. The results on multiple datasets demonstrate that OpenObj achieves superior performance in zero-shot segmentation and retrieval tasks. Additionally, OpenObj supports real-world robotics tasks at several levels, including global movement and local manipulation.

14 October 2025 in IEEE Transactions on Robotics
Verified
Journal Article
Closed access
YDYinan Deng
Yinan Deng
YYYufeng Yue
Yufeng Yue
JDJianyu Dou
Jianyu Dou
JZJingyu Zhao
Jingyu Zhao
JWJiahui Wang
Jiahui Wang
MFMengyin Fu
Mengyin Fu
Robotic systems demand accurate and comprehensive 3D environment perception, requiring simultaneous capture of photo-realistic appearance (optical), precise layout shape (geometric), and open-vocabulary scene understanding (semantic). Existing methods typically achieve only partial fulfillment of these requirements while exhibiting optical blurring, geometric irregularities, and semantic ambiguities. To address these challenges, we propose OmniMap. Overall, OmniMap represents the first online mapping framework that simultaneously captures optical, geometric, and semantic scene attributes while maintaining real-time performance and model compactness. At the architectural level, OmniMap employs a tightly coupled 3DGS-Voxel hybrid representation that combines fine-grained modeling with structural stability. At the implementation level, OmniMap identifies key challenges across different modalities and introduces several innovations: adaptive camera modeling for motion blur and exposure compensation, hybrid incremental representation with normal constraints, and probabilistic fusion for robust instance-level understanding. Extensive experiments show OmniMap's superior performance in rendering fidelity, geometric accuracy, and zero-shot semantic segmentation compared to state-of-the-art methods across diverse scenes. The framework's versatility is further evidenced through a variety of downstream applications, including multi-domain scene Q&A, interactive editing, perception-guided manipulation, and map-assisted navigation.