A Study on Machine Learning-Enhanced Roadside Unit-Based Detection of Abnormal Driving in Autonomous Vehicles

Yun, Keon; Yun, Heesun; Lee, Sangmin; Oh, Jinhyeok; Kim, Minchul; Lim, Myongcheol; Lee, Juntaek; Kim, Chanmin; Seo, Jiwon; Choi, Jinyoung

doi:10.3390/electronics13020288

Cited by 5 publications

(7 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The future work also includes enhancing model capabilities to distinguish between different anomaly types such as short spikes, noise, gradual drift, etc. Being able to characterize the root cause of anomalies is important for autonomous vehicles to take appropriate actions [41]. The model could be extended with techniques such as pattern-based classification on the reconstruction errors to categorize different falsified trajectory behaviors.…”

Section: Discussionmentioning

confidence: 99%

Anomaly Detection in Connected and Autonomous Vehicle Trajectories Using LSTM Autoencoder and Gaussian Mixture Model

Wang,

Li,

Khattak

2024

Electronics

View full text Add to dashboard Cite

Connected and Autonomous Vehicles (CAVs) technology has the potential to transform the transportation system. Although these new technologies have many advantages, the implementation raises significant concerns regarding safety, security, and privacy. Anomalies in sensor data caused by errors or cyberattacks can cause severe accidents. To address the issue, this study proposed an innovative anomaly detection algorithm, namely the LSTM Autoencoder with Gaussian Mixture Model (LAGMM). This model supports anomalous CAV trajectory detection in the real-time leveraging communication capabilities of CAV sensors. The LSTM Autoencoder is applied to generate low-rank representations and reconstruct errors for each input data point, while the Gaussian Mixture Model (GMM) is employed for its strength in density estimation. The proposed model was jointly optimized for the LSTM Autoencoder and GMM simultaneously. The study utilizes realistic CAV data from a platooning experiment conducted for Cooperative Automated Research Mobility Applications (CARMAs). The experiment findings indicate that the proposed LAGMM approach enhances detection accuracy by 3% and precision by 6.4% compared to the existing state-of-the-art methods, suggesting a significant improvement in the field.

show abstract

Section: Discussionmentioning

confidence: 99%

Anomaly Detection in Connected and Autonomous Vehicle Trajectories Using LSTM Autoencoder and Gaussian Mixture Model

Wang,

Li,

Khattak

2024

Electronics

View full text Add to dashboard Cite

show abstract

“…Improving highway safety is one of the most developed areas of ML application in ITSs. It includes security analysis of the road network and its elements [56][57][58], improving information security in ITSs [59,60], and detecting anomalies in sensor data of connected and autonomous vehicles [61][62][63]. It also involves intelligent analysis of road traffic accidents to identify critical factors in their occurrence, essential for preventing similar incidents in the future [64].…”

Section: Machine Learning For Intelligent Transport System Technologiesmentioning

confidence: 99%

Machine Learning-Driven Calibration of Traffic Models Based on a Real-Time Video Analysis

Lopukhova,

Abdulnagimov,

Voronkov

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Accurate traffic simulation models play a crucial role in developing intelligent transport systems that offer timely traffic information to users and efficient traffic management. However, calibrating these models to represent real-world traffic conditions accurately poses a significant challenge due to the dynamic nature of traffic flow and the limitations of traditional calibration methods. This article introduces a machine learning-based approach to calibrate macroscopic traffic simulation models using real-time traffic video stream data. The proposed method for creating and calibrating a traffic simulation model has significantly improved the statistical correspondence between the generated vehicle characteristics and real data about cars on the simulated road section. The correspondence has increased from 37% to 73%. Machine learning models trained on generated data and tested on real data show improved accuracy rates. Mean absolute error, mean square error, and mean absolute percentage error decreased by more than two orders of magnitude. The coefficient of determination has also increased, approaching 1. This method eliminates the need to deploy wireless sensor networks, which can reduce the cost of implementing intelligent transport systems.

show abstract

“…The basic safety message (BSM) [11,14,15] is generated by vehicles and facilitates both vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications, as illustrated in Figure 1. Conversely, probe vehicle data (PVD) [14] originate from Roadside Units (RSUs), serving as infrastructure-to-vehicle (I2V) communication.…”

Section: Introductionmentioning

confidence: 99%

“…2024, 15 This paper discusses the role of the roadside units (RSUs) and how they interact with autonomous vehicles. The RSUs are important components of transportation infrastructure and exchange information with autonomous vehicles through vehicle-to-infrastructure (V2I) communications [10,11]. This communication provides vehicles with important information such as real-time traffic conditions, road conditions, and safety-related warnings, increasing vehicle operating efficiency and improving traffic safety.…”

Section: Introductionmentioning

confidence: 99%

A Study on Reducing Traffic Congestion in the Roadside Unit for Autonomous Vehicles Using BSM and PVD

Lee,

Oh,

Kim

et al. 2024

WEVJ

Self Cite

View full text Add to dashboard Cite

With the rapid advancement of autonomous vehicles reshaping urban transportation, the importance of innovative traffic management solutions has escalated. This research addresses these challenges through the deployment of roadside units (RSUs), aimed at enhancing traffic flow and safety within the autonomous driving era. Our research, conducted in diverse road settings such as straight and traffic circle roads, delves into the RSUs’ capacity to diminish traffic density and alleviate congestion. Employing vehicle-to-infrastructure communication, we can scrutinize its essential role in navigating autonomous vehicles, incorporating basic safety messages (BSMs) and probe vehicle data (PVD) to accurately monitor vehicle presence and status. This paper presupposes the connectivity of all vehicles, contemplating the integration of on-board units or on-board diagnostics in legacy vehicles to extend connectivity, albeit this aspect falls beyond the work’s current ambit. Our detailed experiments on two types of roads demonstrate that vehicle behavior is significantly impacted when density reaches critical thresholds of 3.57% on straight roads and 34.41% on traffic circle roads. However, it is important to note that the identified threshold values are not absolute. In our experiments, these thresholds represent points at which the behavior of one vehicle begins to significantly impact the flow of two or more vehicles. At these levels, we propose that RSUs intervene to mitigate traffic issues by implementing measures such as prohibiting lane changes or restricting entry to traffic circles. We propose a new message set in PVD for RSUs: road balance. Using this message, RSUs can negotiate between vehicles. This approach underscores the RSUs’ capability to actively manage traffic flow and prevent congestion, highlighting their critical role in maintaining optimal traffic conditions and enhancing road safety.

show abstract

A Study on Machine Learning-Enhanced Roadside Unit-Based Detection of Abnormal Driving in Autonomous Vehicles

Cited by 5 publications

References 69 publications

Anomaly Detection in Connected and Autonomous Vehicle Trajectories Using LSTM Autoencoder and Gaussian Mixture Model

Anomaly Detection in Connected and Autonomous Vehicle Trajectories Using LSTM Autoencoder and Gaussian Mixture Model

Machine Learning-Driven Calibration of Traffic Models Based on a Real-Time Video Analysis

A Study on Reducing Traffic Congestion in the Roadside Unit for Autonomous Vehicles Using BSM and PVD

Contact Info

Product

Resources

About