In this paper, we leverage vehicle-to-vehicle (V2V) communication technology to implement computationally efficient decentralized algorithms for two-vehicle cooperative collision avoidance at intersections. Our algorithms employ formal control theoretic methods to guarantee a collision-free (safe) system, whereas overrides are only applied when necessary to prevent a crash. Model uncertainty and communication delays are explicitly accounted for by the model and by the state estimation algorithm. The main contribution of this work is to provide an experimental validation of our method on two instrumented vehicles engaged in an intersection collision avoidance scenario in a test track.

/pdf/cooperative-collision-avoidance-at-intersections-algorithms-2lz2ijuket.pdf

Cooperative Collision Avoidance at Intersections: Algorithms and Experiments

The Internet of Things (IoT) has been growing in recent years with the improvements in several different applications in the military, marine, intelligent transportation, smart health, smart grid, smart home and smart city domains. Although IoT brings significant advantages over traditional information and communication (ICT) technologies for Intelligent Transportation Systems (ITS), these applications are still very rare. Although there is a continuous improvement in road and vehicle safety, as well as improvements in IoT, the road traffic accidents have been increasing over the last decades. Therefore, it is necessary to find an effective way to reduce the frequency and severity of traffic accidents. Hence, this paper presents an intelligent traffic accident detection system in which vehicles exchange their microscopic vehicle variables with each other. The proposed system uses simulated data collected from vehicular ad-hoc networks (VANETs) based on the speeds and coordinates of the vehicles and then, it sends traffic alerts to the drivers. Furthermore, it shows how machine learning methods can be exploited to detect accidents on freeways in ITS. It is shown that if position and velocity values of every vehicle are given, vehicles' behavior could be analyzed and accidents can be detected easily. Supervised machine learning algorithms such as Artificial Neural Networks (ANN), Support Vector Machine (SVM), and Random Forests (RF) are implemented on traffic data to develop a model to distinguish accident cases from normal cases. The performance of RF algorithm, in terms of its accuracy, was found superior to ANN and SVM algorithms. RF algorithm has showed better performance with 91.56% accuracy than SVM with 88.71% and ANN with 90.02% accuracy.

Traffic accident detection using random forest classifier

This paper presents a vision-based real-time gaze zone estimator based on a driver's head orientation composed of yaw and pitch. Generally, vision-based methods are vulnerable to the wearing of eyeglasses and image variations between day and night. The proposed method is novel in the following four ways: First, the proposed method can work under both day and night conditions and is robust to facial image variation caused by eyeglasses because it only requires simple facial features and not specific features such as eyes, lip corners, and facial contours. Second, an ellipsoidal face model is proposed instead of a cylindrical face model to exactly determine a driver's yaw. Third, we propose new features-the normalized mean and the standard deviation of the horizontal edge projection histogram-to reliably and rapidly estimate a driver's pitch. Fourth, the proposed method obtains an accurate gaze zone by using a support vector machine. Experimental results from 200 000 images showed that the root mean square errors of the estimated yaw and pitch angles are below 7 under both daylight and nighttime conditions. Equivalent results were obtained for drivers with glasses or sunglasses, and 18 gaze zones were accurately estimated using the proposed gaze estimation method.

https://web.yonsei.ac.kr/hgjung/Ho%20Gi%20Jung%20Homepage/Publications/2011/T-ITS(Real-Time%20Gaze%20Estimator%20Based%20on%20Driver’s%20Head%20Orientation%20for%20Forward%20Collision%20Warning%20System).pdf

Real-Time Gaze Estimator Based on Driver's Head Orientation for Forward Collision Warning System

In vehicular ad hoc networks (VANETs), cooperation between nodes is needed for proper performance of safety and traffic efficiency applications A non-trivial scientific challenge in VANETs is the design of an adaptive broadcast protocol, which can provide efficient and end-to-end reliable warning messages dissemination Usually, broadcast protocols for VANETs use beacon messages, disseminated among the vehicles, in order to get neighborhood information When the vehicles are next to each other trying to broadcast at the same time, this may lead to frequent contention, and broadcast storms On the other hand, in sparse density scenarios, vehicles have to face with failures in the message delivery In this paper, we propose a novel protocol, Adaptive Data Dissemination Protocol (AddP), which aims at providing reliability to message dissemination in an efficient way The protocol uses different mechanisms to dynamically adjust the beacon periodicity and to reduce the number of messages and beacons in the network The effectiveness of our proposed protocol and the impacts of the use of the protocol by a VANET safety application were evaluated through experiments with network and traffic simulators

Reliable data dissemination protocol for VANET traffic safety applications

This paper proposes a novel anomaly detection and classification algorithm that combines the spatiotemporal changes in the variability of microscopic traffic variables, namely, relative speed, intervehicle time gap, and lane changing. When applied to real-world scenarios, the proposed algorithm can use the variances of statistics of microscopic traffic variables to detect and classify traffic anomalies. Based on a simulation environment, it is shown that, with minimum prior knowledge and partial availability of microscopic traffic information from as few as 20% of the vehicle population, the proposed algorithm can still achieve 100% detection rates and very low false alarm rates, which outperforms previous algorithms monitoring loop detectors that are ideally placed at locations where anomalies originate.

Detection and Classification of Traffic Anomalies Using Microscopic Traffic Variables

Vehicle collision mitigation, cooperative driving, and vehicle-to-vehicle (V2V) and/or vehicle-to-infrastructure (V2I) communication constitute a broad multidisciplinary research field that focuses on improving road safety. Statistics indicate that the primary cause of most road accidents is vehicles' excessive speed and delayed drivers reaction. Thus, road safety can be improved by early warning based on V2V communication. An innovative system called wireless local danger warning (WILLWARN), which is based on recent and future trends of cooperative driving, enables an electronic safety horizon for foresighted driving by implementing onboard vehicle-hazard detection and V2V communication. One of the key innovative features of the proposed system is the focus on low penetration levels in rural traffic by a new message-management strategy that is based on storing warning information in the vehicle and distributing warnings through communication, particularly with oncoming traffic. The system timely warns the driver about a dangerous situation ahead by decentralized distribution of warnings and incident messages via ad hoc intervehicle communication. The WILLWARN system is based on a modular object-oriented architecture consisting of the V2V communication module (VVC), the warning message-management module (WMM), the hazard-detection-management module (HDM), the hazard-warning-management module (HWM), a Global Positioning System (GPS) receiver, and various onboard sensors. In this paper, all system modules, as well as their interoperability, are presented in detail.

F Aguado-Agelet

Papers

Wireless Local Danger Warning: Cooperative Foresighted Driving Using Intervehicle Communication