Video driven traffic modelling

We propose Video Driven Traffic Modelling (VDTM) for accurate simulation of real-world traffic behaviours with detailed information and low-cost model development and maintenance. Computer vision techniques are employed to estimate traffic parameters. These parameters are used to build and update a traffic system model. The model is simulated using the Paramics traffic simulation platform. Based on the simulation techniques, effects of traffic interventions can be evaluated in order to achieve better decision makings for traffic management authorities. In this paper, traffic parameters such as vehicle types, times of starting trips and corresponding origin-destinations are extracted from a video. A road network is manually defined according to the traffic composition in the video, and individual vehicles associated with extracted properties are modelled and simulated within the defined road network using Paramics. VDTM has widespread potential applications in supporting traffic decision-makings. To demonstrate the effectiveness, we apply it in optimizing a traffic signal control system, which adaptively adjusts green times of signals at an intersection to reduce traffic congestion.

2Loud? Monitoring traffic noise with mobile phones

The World Health Organization has recently focused attention on guidelines for night noise in urban areas, based on significant medical evidence of the adverse impacts of exposure to excessive traffic noise on health, especially caused by sleep disturbance. This includes serious illnesses, such as hypertension, arteriosclerosis and myocardial infarction. 2Loud? is a research project with the aim of developing and testing a mobile phone application to allow a community to monitor traffic noise in their environment, with focus on the night period and indoor measurement. Individuals, using mobile phones, provide data on characteristics of their dwellings and systematically record the level of noise inside their homes overnight. The records from multiple individuals are sent to a server, integrated into indicators and shared through mapping. The 2Loud? application is not designed to replace existing scientific measurements, but to add information which is currently not available. Noise measurements to assist the planning and management of traffic noise are normally carried out by designated technicians, using sophisticated equipment, and following specific guidelines for outdoors locations. This process provides very accurate records, however, for being a time consuming and expensive system, it results in a limited number of locations being surveyed and long time between updates. Moreover, scientific noise measurements do not survey inside dwellings. In this paper we present and discuss the participatory process proposed, and currently under implementation and test, to characterize the levels of exposure to traffic noise of residents living in the vicinity of highways in the City of Boroondara (Victoria, Australia) using the 2Loud? application.

Drivers display anger-congruent attention to potential traffic hazards

Previous research has suggested that angry drivers may respond differently to potential hazards. This study replicates and extends these findings. Under simulated driving conditions, two groups of drivers experienced conditions that would either increase angry mood (N=12; men =6) or not (control group, N =12; men=6). All drivers then performed a neutral drive, during which they encountered a number of traffic events not experienced in the initial drive. These included vehicles emerging from driveways into their path and jaywalking pedestrians. Subjective anger, eye-movement behaviour and driving behaviours (speed and reaction times) were measured as drivers drove. Subjective moods (Profile of Mood States) were assessed before and after each drive. Anger-provoked drivers reported reliably higher increases in angry mood when compared with the control group after the initial drive, and these increases remained stable across the subsequent neutral drive. During the neutral drive, anger provoked drivers demonstrated evidence of more heuristic style processing of potential hazards, with shorter initial gazes at less apparent hazards and longer latencies to look back at jaywalking pedestrians obscured by parked vehicles. Anger-provoked drivers also took longer to make corrective actions to avoid potential collisions. It is concluded that anger-provoked drivers may initially make more superficial assessments of certain driving situations and consequently underestimate the inherent risk.

Intelligent traffic light control of isolated intersections using machine learning methods

Traffic congestion is one of the major problems in modern cities. This study applies machine learning methods to determine green times in order to minimize in an isolated intersection. Q-learning and neural networks are applied here to set signal light times and minimize total delays. It is assumed that an intersection behaves in a similar fashion to an intelligent agent learning how to set green times in each cycle based on traffic information. Here, a comparison between Q-learning and neural network is presented. In Q-learning, considering continuous green time requires a large state space, making the learning process practically impossible. In contrast to Q-learning methods, the neural network model can easily set the appropriate green time to fit the traffic demand. The performance of the proposed neural network is compared with two traditional alternatives for controlling traffic lights. Simulation results indicate that the application of the proposed method greatly reduces the total delay in the network compared to the alternative methods.

Network traffic clustering using random forest proximities

The recent years have seen extensive work on statistics-based network traffic classification using machine learning (ML) techniques. In the particular scenario of learning from unlabeled traffic data, some classic unsupervised clustering algorithms (e.g. K-Means and EM) have been applied but the reported results are unsatisfactory in terms of low accuracy. This paper presents a novel approach for the task, which performs clustering based on Random Forest (RF) proximities instead of Euclidean distances. The approach consists of two steps. In the first step, we derive a proximity measure for each pair of data points by performing a RF classification on the original data and a set of synthetic data. In the next step, we perform a K-Medoids clustering to partition the data points into K groups based on the proximity matrix. Evaluations have been conducted on real-world Internet traffic traces and the experimental results indicate that the proposed approach is more accurate than the previous methods.

Classifying network traffic in the big data era

With the arrival of Big Data Era, properly utilizing the power of big data is becoming increasingly essential for the strength and competitiveness of businesses and organizations. We are facing grand challenges from big data from different perspectives, such as processing, communication, security, and privacy. In this talk, we discuss the big data challenges in network traffic classification and our solutions to the challenges. The significance of the research lies in the fact that each year the network traffic increase exponentially on the current Internet. Traffic classification has wide applications in network management, from security monitoring to quality of service measurements. Recent research tends to apply machine-learning techniques to flow statistical feature based classification methods. In this talk, we propose a series of novel approaches for traffic classification, which can improve the classification performance effectively by incorporating correlated information into the classification process. We analyze the new classification approaches and their performance benefit from both theoretical and empirical perspectives. A large number of experiments are carried out on two real-world traffic datasets to validate the proposed approach. The results show the traffic classification performance can be improved significantly even under the extreme difficult circumstance of very few training samples. Our work has significant impact on security applications.

Estimating fatality rates in occupational light vehicle users using vehicle registration and crash data

Objective: To estimate occupational light vehicle (OLV) fatality numbers using vehicle registration and crash data and compare these with previous estimates based on workers' compensation data. Method: New South Wales (NSW) Roads and Traffic Authority (RTA) vehicle registration and crash data were obtained for 2004. NSW is the only Australian jurisdiction with mandatory work-use registration, which was used as a proxy for work-relatedness. OLV fatality rates based on registration data as the denominator were calculated and comparisons made with published 2003/04 fatalities based on workers' compensation data. Results: Thirty-four NSW RTA OLV-user fatalities were identified, a rate of 4.5 deaths per 100,000 organisationally registered OLV, whereas the Australian Safety and Compensation Council (ASCC), reported 28 OLV deaths Australia-wide. Conclusions: More OLV user fatalities were identified from vehicle registration-based data than those based on workers' compensation estimates and the data are likely to provide an improved estimate of fatalities specific to OLV use. Implications: OLV-use is an important cause of traumatic fatalities that would be better identified through the use of vehicle-registration data, which provides a stronger evidence base from which to develop policy responses.

Automatic network traffic classification

The thesis addresses a number of critical problems in regard to fully automating the process of network traffic classification and protocol identification. Several effective solutions based on statistical analysis and machine learning techniques are proposed, which significantly reduce the requirements for human interventions in network traffic classification systems.