Weining Lu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Weining Lu has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Bin Liang
Tsinghua University
5 shared publications Bin Liang
Beijing National Research Center for Information Science and Technology Tsinghua University 100084 Bejing China
4 shared publications- Yu Cheng
The Chinese University of Hong Kong, Hong Kong, China
4 shared publications - Pan Zhou
Huazhong University of Science and Technology
3 shared publications - Qingquan Lin
Department of Automation, Tsinghua University, Beijing, China
3 shared publications - Chenxi Li
Department of Automation, Tsinghua University, Beijing, China
2 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (21) , 33103-33110
Learning-Aided Wheel-Encoder Reliability Assessment and Adaptive Sensor Fusion in Ground-Robot Odometry
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (21) , 33103-33110
Learning-Aided Wheel-Encoder Reliability Assessment and Adaptive Sensor Fusion in Ground-Robot Odometry
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 30 (6) , 4104-4115
SEE : A New Benchmark and Baseline for S afe, E fficient, and E xtensive Off-Road Exploration
Autonomous exploration in off-road environments faces significant challenges regarding safety and exploration coverage. Current sampling-based approaches do not sufficiently address off-road safety considerations, while optimization-based methods fail to achieve adequate coverage in open environments. We address this with a risk-heuristic hybrid framework integrating sampling-based local planning and optimization-based global planning. Key innovations include the following. First, a risk map that models safety uncertainties via physical-aware conditional value at risk, combining geometric and physical risks. Second, a risk-heuristic planner that directs sampling toward safe, high-reward trajectories. Third, a hybrid strategy, which enables reliable trajectory generation in off-road environments while maintaining exploration efficiency. Lastly, a benchmark with novel metrics, including slip accumulation derived from proprioceptive state estimation discrepancies. Evaluations in simulation and real-world trials show our method achieves 58.4% larger exploration coverage and 67.1% lower slip accumulation versus state-of-the-art approaches. The framework enables robust navigation in challenging terrains where traditional optimization-based planners fail, advancing autonomy in safety-critical environments.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 30 (6) , 4104-4115
SEE : A New Benchmark and Baseline for S afe, E fficient, and E xtensive Off-Road Exploration
Autonomous exploration in off-road environments faces significant challenges regarding safety and exploration coverage. Current sampling-based approaches do not sufficiently address off-road safety considerations, while optimization-based methods fail to achieve adequate coverage in open environments. We address this with a risk-heuristic hybrid framework integrating sampling-based local planning and optimization-based global planning. Key innovations include the following. First, a risk map that models safety uncertainties via physical-aware conditional value at risk, combining geometric and physical risks. Second, a risk-heuristic planner that directs sampling toward safe, high-reward trajectories. Third, a hybrid strategy, which enables reliable trajectory generation in off-road environments while maintaining exploration efficiency. Lastly, a benchmark with novel metrics, including slip accumulation derived from proprioceptive state estimation discrepancies. Evaluations in simulation and real-world trials show our method achieves 58.4% larger exploration coverage and 67.1% lower slip accumulation versus state-of-the-art approaches. The framework enables robust navigation in challenging terrains where traditional optimization-based planners fail, advancing autonomy in safety-critical environments.
- 29 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 20702-20712
Nav-SCOPE: Swarm Robot Cooperative Perception and Coordinated Navigation
This paper proposes a lightweight decentralized solution for multi-robot coordinated navigation with cooperative perception. First, we introduce a rapid way to process sensory data, thus obtaining safe directions and key environmental features. Then, an information flow is created to facilitate real-time perception sharing over wireless ad-hoc networks. Consequently, the environmental uncertainties of each robot are reduced by interaction fields that deliver complementary information. Finally, path optimization is achieved in a probabilistic way, enabling self-organized coordination with effective convergence, divergence, and collision avoidance. Our method is fully interpretable and ready for deployment without gaps. Comprehensive simulations and real-world experiments demonstrate reduced path redundancy, robust performance across various tasks, and minimal demands on computation and communication.
- 29 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 20702-20712
Nav-SCOPE: Swarm Robot Cooperative Perception and Coordinated Navigation
This paper proposes a lightweight decentralized solution for multi-robot coordinated navigation with cooperative perception. First, we introduce a rapid way to process sensory data, thus obtaining safe directions and key environmental features. Then, an information flow is created to facilitate real-time perception sharing over wireless ad-hoc networks. Consequently, the environmental uncertainties of each robot are reduced by interaction fields that deliver complementary information. Finally, path optimization is achieved in a probabilistic way, enabling self-organized coordination with effective convergence, divergence, and collision avoidance. Our method is fully interpretable and ready for deployment without gaps. Comprehensive simulations and real-world experiments demonstrate reduced path redundancy, robust performance across various tasks, and minimal demands on computation and communication.
- 19 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (12) , 13617-13626
Collaborative Navigation of Multiple Autonomous Mobile Robots via Geometric Graph Neural Network
Multiple mobile robots play a crucial role in spatially distributed tasks. In unknown and nonrepetitive scenarios, reconstructing a global map is time-consuming and often unnecessary. Hence, research has focused on real-time collaborative planning without relying on a global map. This article introduces a novel multirobot collaborative path planning method based on geometric graph neural network (MRPP-GeoGNN). With all robots connected via local communication devices, the features of neighboring robot’s sensory data are transmitted to the ego-robot for reference. The GeoGNN layers acquire local surrounding features by integrating these sensory features with positional embeddings. We use a geometric function to project positional information into a high-dimensional space, which ensures the rotation invariance of different nodes in a relative coordinate system. An action mapper finally maps the integrated feature to multiple forward directions to guide the robot’s movement. Simulation results demonstrate an approximate 5% improvement in accuracy on expert datasets compared to convolutional neural network (CNN), along with a 4% increase in success rate and an 18% reduction in flowtime increase in robot operating system (ROS) tests, outperforming other GNN models. Real-world experiments further validate its ability to effectively leverage neighbor information, significantly improving path efficiency in practical applications.
- 19 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (12) , 13617-13626
Collaborative Navigation of Multiple Autonomous Mobile Robots via Geometric Graph Neural Network
Multiple mobile robots play a crucial role in spatially distributed tasks. In unknown and nonrepetitive scenarios, reconstructing a global map is time-consuming and often unnecessary. Hence, research has focused on real-time collaborative planning without relying on a global map. This article introduces a novel multirobot collaborative path planning method based on geometric graph neural network (MRPP-GeoGNN). With all robots connected via local communication devices, the features of neighboring robot’s sensory data are transmitted to the ego-robot for reference. The GeoGNN layers acquire local surrounding features by integrating these sensory features with positional embeddings. We use a geometric function to project positional information into a high-dimensional space, which ensures the rotation invariance of different nodes in a relative coordinate system. An action mapper finally maps the integrated feature to multiple forward directions to guide the robot’s movement. Simulation results demonstrate an approximate 5% improvement in accuracy on expert datasets compared to convolutional neural network (CNN), along with a 4% increase in success rate and an 18% reduction in flowtime increase in robot operating system (ROS) tests, outperforming other GNN models. Real-world experiments further validate its ability to effectively leverage neighbor information, significantly improving path efficiency in practical applications.
- Open Access
Efficient Collaborative Navigation through Perception Fusion for Multi-Robots in Unknown Environments
For tasks conducted in unknown environments with efficiency requirements, real-time navigation of multi-robot systems remains challenging due to unfamiliarity with surroundings.In this paper, we propose a novel multi-robot collaborative planning method that leverages the perception of different robots to intelligently select search directions and improve planning efficiency. Specifically, a foundational planner is employed to ensure reliable exploration towards targets in unknown environments and we introduce Graph Attention Architecture with Information Gain Weight(GIWT) to synthesizes the information from the target robot and its teammates to facilitate effective navigation around obstacles.In GIWT, after regionally encoding the relative positions of the robots along with their perceptual features, we compute the shared attention scores and incorporate the information gain obtained from neighboring robots as a supplementary weight. We design a corresponding expert data generation scheme to simulate real-world decision-making conditions for network training. Simulation experiments and real robot tests demonstrates that the proposed method significantly improves efficiency and enables collaborative planning for multiple robots. Our method achieves approximately 82% accuracy on the expert dataset and reduces the average path length by about 8% and 6% across two types of tasks compared to the fundamental planner in ROS tests, and a path length reduction of over 6% in real-world experiments.
Open AccessEfficient Collaborative Navigation through Perception Fusion for Multi-Robots in Unknown Environments
For tasks conducted in unknown environments with efficiency requirements, real-time navigation of multi-robot systems remains challenging due to unfamiliarity with surroundings.In this paper, we propose a novel multi-robot collaborative planning method that leverages the perception of different robots to intelligently select search directions and improve planning efficiency. Specifically, a foundational planner is employed to ensure reliable exploration towards targets in unknown environments and we introduce Graph Attention Architecture with Information Gain Weight(GIWT) to synthesizes the information from the target robot and its teammates to facilitate effective navigation around obstacles.In GIWT, after regionally encoding the relative positions of the robots along with their perceptual features, we compute the shared attention scores and incorporate the information gain obtained from neighboring robots as a supplementary weight. We design a corresponding expert data generation scheme to simulate real-world decision-making conditions for network training. Simulation experiments and real robot tests demonstrates that the proposed method significantly improves efficiency and enables collaborative planning for multiple robots. Our method achieves approximately 82% accuracy on the expert dataset and reduces the average path length by about 8% and 6% across two types of tasks compared to the fundamental planner in ROS tests, and a path length reduction of over 6% in real-world experiments.
- Open Access
- 25 July 2024
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (5)
CatLearning: highly accurate gene expression prediction from histone mark
Histone modifications, known as histone marks, are pivotal in regulating gene expression within cells. The vast array of potential combinations of histone marks presents a considerable challenge in decoding the regulatory mechanisms solely through biological experimental approaches. To overcome this challenge, we have developed a method called CatLearning. It utilizes a modified convolutional neural network architecture with a specialized adaptation Residual Network to quantitatively interpret histone marks and predict gene expression. This architecture integrates long-range histone information up to 500Kb and learns chromatin interaction features without 3D information. By using only one histone mark, CatLearning achieves a high level of accuracy. Furthermore, CatLearning predicts gene expression by simulating changes in histone modifications at enhancers and throughout the genome. These findings help comprehend the architecture of histone marks and develop diagnostic and therapeutic targets for diseases with epigenetic changes.
- 25 July 2024
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (5)
Open AccessCatLearning: highly accurate gene expression prediction from histone mark
Histone modifications, known as histone marks, are pivotal in regulating gene expression within cells. The vast array of potential combinations of histone marks presents a considerable challenge in decoding the regulatory mechanisms solely through biological experimental approaches. To overcome this challenge, we have developed a method called CatLearning. It utilizes a modified convolutional neural network architecture with a specialized adaptation Residual Network to quantitatively interpret histone marks and predict gene expression. This architecture integrates long-range histone information up to 500Kb and learns chromatin interaction features without 3D information. By using only one histone mark, CatLearning achieves a high level of accuracy. Furthermore, CatLearning predicts gene expression by simulating changes in histone modifications at enhancers and throughout the genome. These findings help comprehend the architecture of histone marks and develop diagnostic and therapeutic targets for diseases with epigenetic changes.
- Open Access
- 11 January 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 20 (4) , 1-19
Transform-Equivariant Consistency Learning for Temporal Sentence Grounding
This paper addresses the temporal sentence grounding (TSG). Although existing methods have made decent achievements in this task, they not only severely rely on abundant video-query paired data for training, but also easily fail into the dataset distribution bias. To alleviate these limitations, we introduce a novel Equivariant Consistency Regulation Learning (ECRL) framework to learn more discriminative query-related frame-wise representations for each video, in a self-supervised manner. Our motivation comes from that the temporal boundary of the query-guided activity should be consistently predicted under various video-level transformations. Concretely, we first design a series of spatio-temporal augmentations on both foreground and background video segments to generate a set of synthetic video samples. In particular, we devise a self-refine module to enhance the completeness and smoothness of the augmented video. Then, we present a novel self-supervised consistency loss (SSCL) applied on the original and augmented videos to capture their invariant query-related semantic by minimizing the KL-divergence between the sequence similarity of two videos and a prior Gaussian distribution of timestamp distance. At last, a shared grounding head is introduced to predict the transform-equivariant query-guided segment boundaries for both the original and augmented videos. Extensive experiments on three challenging datasets (ActivityNet, TACoS, and Charades-STA) demonstrate both effectiveness and efficiency of our proposed ECRL framework.
- 11 January 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 20 (4) , 1-19
Open AccessTransform-Equivariant Consistency Learning for Temporal Sentence Grounding
This paper addresses the temporal sentence grounding (TSG). Although existing methods have made decent achievements in this task, they not only severely rely on abundant video-query paired data for training, but also easily fail into the dataset distribution bias. To alleviate these limitations, we introduce a novel Equivariant Consistency Regulation Learning (ECRL) framework to learn more discriminative query-related frame-wise representations for each video, in a self-supervised manner. Our motivation comes from that the temporal boundary of the query-guided activity should be consistently predicted under various video-level transformations. Concretely, we first design a series of spatio-temporal augmentations on both foreground and background video segments to generate a set of synthetic video samples. In particular, we devise a self-refine module to enhance the completeness and smoothness of the augmented video. Then, we present a novel self-supervised consistency loss (SSCL) applied on the original and augmented videos to capture their invariant query-related semantic by minimizing the KL-divergence between the sequence similarity of two videos and a prior Gaussian distribution of timestamp distance. At last, a shared grounding head is introduced to predict the transform-equivariant query-guided segment boundaries for both the original and augmented videos. Extensive experiments on three challenging datasets (ActivityNet, TACoS, and Charades-STA) demonstrate both effectiveness and efficiency of our proposed ECRL framework.
- Open Access
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (2) , 1640-1648
Hypotheses Tree Building for One-Shot Temporal Sentence Localization
Given an untrimmed video, temporal sentence localization (TSL) aims to localize a specific segment according to a given sentence query. Though respectable works have made decent achievements in this task, they severely rely on dense video frame annotations, which require a tremendous amount of human effort to collect. In this paper, we target another more practical and challenging setting: one-shot temporal sentence localization (one-shot TSL), which learns to retrieve the query information among the entire video with only one annotated frame. Particularly, we propose an effective and novel tree-structure baseline for one-shot TSL, called Multiple Hypotheses Segment Tree (MHST), to capture the query-aware discriminative frame-wise information under the insufficient annotations. Each video frame is taken as the leaf-node, and the adjacent frames sharing the same visual-linguistic semantics will be merged into the upper non-leaf node for tree building. At last, each root node is an individual segment hypothesis containing the consecutive frames of its leaf-nodes. During the tree construction, we also introduce a pruning strategy to eliminate the interference of query-irrelevant nodes. With our designed self-supervised loss functions, our MHST is able to generate high-quality segment hypotheses for ranking and selection with the query. Experiments on two challenging datasets demonstrate that MHST achieves competitive performance compared to existing methods.
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (2) , 1640-1648
Open AccessHypotheses Tree Building for One-Shot Temporal Sentence Localization
Given an untrimmed video, temporal sentence localization (TSL) aims to localize a specific segment according to a given sentence query. Though respectable works have made decent achievements in this task, they severely rely on dense video frame annotations, which require a tremendous amount of human effort to collect. In this paper, we target another more practical and challenging setting: one-shot temporal sentence localization (one-shot TSL), which learns to retrieve the query information among the entire video with only one annotated frame. Particularly, we propose an effective and novel tree-structure baseline for one-shot TSL, called Multiple Hypotheses Segment Tree (MHST), to capture the query-aware discriminative frame-wise information under the insufficient annotations. Each video frame is taken as the leaf-node, and the adjacent frames sharing the same visual-linguistic semantics will be merged into the upper non-leaf node for tree building. At last, each root node is an individual segment hypothesis containing the consecutive frames of its leaf-nodes. During the tree construction, we also introduce a pruning strategy to eliminate the interference of query-irrelevant nodes. With our designed self-supervised loss functions, our MHST is able to generate high-quality segment hypotheses for ranking and selection with the query. Experiments on two challenging datasets demonstrate that MHST achieves competitive performance compared to existing methods.
- Open Access
Hypotheses Tree Building for One-Shot Temporal Sentence Localization
Given an untrimmed video, temporal sentence localization (TSL) aims to localize a specific segment according to a given sentence query. Though respectable works have made decent achievements in this task, they severely rely on dense video frame annotations, which require a tremendous amount of human effort to collect. In this paper, we target another more practical and challenging setting: one-shot temporal sentence localization (one-shot TSL), which learns to retrieve the query information among the entire video with only one annotated frame. Particularly, we propose an effective and novel tree-structure baseline for one-shot TSL, called Multiple Hypotheses Segment Tree (MHST), to capture the query-aware discriminative frame-wise information under the insufficient annotations. Each video frame is taken as the leaf-node, and the adjacent frames sharing the same visual-linguistic semantics will be merged into the upper non-leaf node for tree building. At last, each root node is an individual segment hypothesis containing the consecutive frames of its leaf-nodes. During the tree construction, we also introduce a pruning strategy to eliminate the interference of query-irrelevant nodes. With our designed self-supervised loss functions, our MHST is able to generate high-quality segment hypotheses for ranking and selection with the query. Experiments on two challenging datasets demonstrate that MHST achieves competitive performance compared to existing methods.
Open AccessHypotheses Tree Building for One-Shot Temporal Sentence Localization
Given an untrimmed video, temporal sentence localization (TSL) aims to localize a specific segment according to a given sentence query. Though respectable works have made decent achievements in this task, they severely rely on dense video frame annotations, which require a tremendous amount of human effort to collect. In this paper, we target another more practical and challenging setting: one-shot temporal sentence localization (one-shot TSL), which learns to retrieve the query information among the entire video with only one annotated frame. Particularly, we propose an effective and novel tree-structure baseline for one-shot TSL, called Multiple Hypotheses Segment Tree (MHST), to capture the query-aware discriminative frame-wise information under the insufficient annotations. Each video frame is taken as the leaf-node, and the adjacent frames sharing the same visual-linguistic semantics will be merged into the upper non-leaf node for tree building. At last, each root node is an individual segment hypothesis containing the consecutive frames of its leaf-nodes. During the tree construction, we also introduce a pruning strategy to eliminate the interference of query-irrelevant nodes. With our designed self-supervised loss functions, our MHST is able to generate high-quality segment hypotheses for ranking and selection with the query. Experiments on two challenging datasets demonstrate that MHST achieves competitive performance compared to existing methods.