Ruoheng Li

Ruoheng Li

Auto-generated profile, based on publicly available information

Ruoheng Li

School of Electronic and Information Engineering, Beihang University, Beijing, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
13 January 2025 in Sustainable Cities and Society
Verified
Journal Article
Closed access
CSChengcheng Song
Chengcheng Song
JYJingjing Yang
Jingjing Yang
ZWZhiyuan Wang
Zhiyuan Wang
RLRuoheng Li
Ruoheng Li
XPXiufeng Pang
Xiufeng Pang
Yixing ChenYixing Chen
Yixing Chen

15 December 2022 in Drones
Verified
Journal Article
Open access
PDF file
SQShao Quan
Shao Quan
JLJiaming Li
Jiaming Li
RLRuoheng Li
Ruoheng Li
JZJiangao Zhang
Jiangao Zhang
XGXiaobo Gao
Xiaobo Gao
The application of drones provides a powerful solution for “the last-mile” logistics services, while the large-scale implementation of logistics drone services will threaten the safety of buildings, pedestrians, vehicles, and other elements in the urban environment. The balance of risk cost and service benefit is accordingly crucial to managing logistics drones. In this study, we proposed a cost-benefit assessment model for quantifying risk cost and service benefit in the urban environment. In addition, a global heuristic path search rule was developed to solve the path planning problem based on risk mitigation and customer service. The cost-benefit assessment model quantifies the risk cost from three environmental elements (buildings, pedestrians, and vehicles) threatened by drone operations based on the collision probability, and the service benefit based on the characteristics of logistics service customers. To explore the effectiveness of the model in this paper, we simulate and analyse the effects of different risk combinations, unknown risk zones, and risk-benefit preferences on the path planning results. The results show that compared with the traditional shortest-distance method, the drone path planning method proposed in this paper can accurately capture the distribution of risks and customers in the urban environment. It is highly reusable in ensuring service benefits while reducing risk costs and generating a cost-effective path for logistics drones. We also compare the algorithm in this paper with the A* algorithm and verify that our algorithm improves the solution quality in complex environments.

10 November 2022 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
SQShao Quan
Shao Quan
RLRuoheng Li
Ruoheng Li
MDMin Dong
Min Dong
CSChengcheng Song
Chengcheng Song
The emerging usage of quadcopters in Urban Air Mobility has urged airspace design to be precise and dynamic to mitigate collision risks. This paper proposes an adaptive AirMatrix model to obtain the block size of AirMatrix that is calculated through the acceptable track deviation of quadcopters by considering GPS signal quality and wind field. First, the acceptable GPS-induced deviation is calculated to meet the safety tolerance by analyzing the distributions of GPS-induced deviation under different signal quality levels. Second, a dynamic modeling and simulation method is applied to assess the wind impact on the track deviation accurately. The wind-induced deviation is calculated using a wind effect model considering surface friction and turbulence. Results show that the proposed model can reduce airspace conserveness by providing 45 percent available blocks more than the previous model. By considering the wind impact, the adaptive AirMatrix improves the safety performance of blocks at levels ranging from 4.9% to 95.5%, compared with the model without considering wind impact. The adaptive AirMatrix model provides a delicate and targeted approach to design airspace for quadcopters, which benefits traffic management in Urban Air Mobility.

3 March 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
RLRuoheng Li
Ruoheng Li
ZLZhongyao Liu
Zhongyao Liu
XZXi Zhu
Xi Zhu
LLLin Li
Lin Li
XCXianbin Cao
Xianbin Cao
Multivariate time series anomaly detection is crucial in the sensitive domains such as cybersecurity and grid monitoring, significantly contributing to reliability and safety of system operation. However, current methods suffer from inadequate utilization of decomposed time series, insufficient mining of contextual dependencies within the time series, and limited robustness against anomalies during training. To address these limitations, we propose ConFlow (Consistencyenhanced normalizing Flow model), which utilizes consistency of decomposed time series and contextual temporal embedding to enhance discriminative ability of flow model. First, to refine extraction of time series components, we propose a cascade decomposition and mixing module that iteratively decouple the time series. Second, these components are mapped to Gaussian distributions through the context-aware normalizing flow, incorporating both inter- and intra-series information into the density estimation. Third, the density consistency among decomposed time series is measured to reweight the estimation, while highly inconsistent series are viewed as anomalies and masked during training to improve model robustness. Finally, anomalies are detected using reweight density estimation. Experiments on five widely used datasets in the time series anomaly detection field demonstrate the superiority of our method over state-of-the-art approaches.

21 April 2026 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
RLRuoheng Li
Ruoheng Li
XLXuhui Liu
Xuhui Liu
YHYutao Hu
Yutao Hu
XZXi Zhu
Xi Zhu
XCXianbin Cao
Xianbin Cao
Drones have become increasingly widely applied in surveillance systems due to their mobility, making aerial video anomaly detection methods more crucial. Anomalies in aerial videos often present as semantic conflicts, such as the presence of unexpected objects or unusual behaviors that do not align with the context. Previous approaches often relied on manually crafted knowledge graphs to detect such conflicts, which suffer from poor scalability. Recently, owing to their sufficient alignment training, multimodal large language models (MLLMs) have emerged as a generalized solution for semantic understanding. However, the direct application of MLLMs does not yield satisfactory anomaly detection performance in aerial videos. First, aerial videos often manifest platform-induced pseudo-motion, which obscures the true motion of objects and exacerbates detection errors. Second, without sufficient labeled data for fine-tuning, generic MLLMs often lack scene-level semantic guidance to reliably distinguish abnormal events that include contextually inappropriate behaviors. To address these challenges, we propose SemAero, an MLLM-based framework to address these challenges by: 1) designing an ego-motion reduction module to enhance model perception on object movement, 2) generating scene-specific prompts adaptively with step-by-step guidance for reasonable output, and 3) refining scores with dual-stream consistent feature for better domain-specific anomaly detection. Evaluated across 8 diverse aerial scenes and 73 sub-datasets, SemAero achieves a 3.09% improvement in AUC-ROC over the second-best model, demonstrating its ability in aerial video anomaly detection.

6 March 2026 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
RLRuoheng Li
Ruoheng Li
TDTang Diyin
Tang Diyin
FWFei Wang
Fei Wang
XZXi Zhu
Xi Zhu
XCXianbin Cao
Xianbin Cao
Timely detection of anomalies in traffic systems is crucial for mitigating risks and economic losses. Current time series anomaly detection methods often use reconstruction errors to identify anomalies, and their accuracy depends on how well they can reconstruct normal patterns from the original series. However, traffic data typically exhibit high noise, spatiotemporal heterogeneity, and uncertain inter-series correlations, complicating the learning of normal patterns. To address these challenges, we introduce SpectraBayes, which explores the reconstruction of density and volume series in the frequency domain for anomaly detection. First, we transform the series into the frequency domain and apply a low-pass filter to remove noise. Then, we embed periodic information into the frequency-domain representation through phase shifts to enhance the temporal awareness. Additionally, we model the inter-series correlations between density and volume resiliently using cross-spectrum probabilistic modeling. Optimized by maximizing the Evidence Lower Bound (ELBO), SpectraBayes ensures robust reconstruction while avoiding overfitting against the uncertain data. SpectraBayes outperforms 21 existing anomaly detection models on traffic series anomaly detection tasks, achieving mean improvements of 2.71% across three metrics over the second-best model. Furthermore, it is lightweight and maintains robust performance under varying noise levels.