Traffic flow dynamics is an important issue for implementing effective pollutant discharge control of tunnels. Longitudinal ventilation using jet fans is the most popular system for pollutant discharge control of tunnels. Nowadays, jet fans equipped with the frequency conversion technology in the tunnel can shorten the control cycle and even conduct manipulation of step-less jet speeds. The longitudinal ventilation system has considerable inertia and non-linear characteristics, which are partly resulted from traffic flow dynamics such as traffic state transition. Therefore in this study an adaptive control method based on the artificial neural-network theory is proposed to be tailored to the traffic state transition. The model is based on aerodynamic equations and takes vehicle speed and density as main system disturbances, whose value can be determined by fundamental diagram when having incomplete field traffic data. The proposed controller can also cope with the parameters and uncertainties of the time-varying model. The authors simulation results show that the adaptive control method can track the desirable system output effectively whenever the traffic condition changes gently or dramatically. The results also show that their method performs better than the common-used proportional integral derivative controller in terms of system adaptability following the traffic state transition.

Accurately understanding driving behaviour is of crucial importance for advanced driving assistant systems such as adaptive cruise control system and intelligent forward collision warning system. To understand different driving styles, this study employs the clustering method and topic model to extract latent driving states, which can elaborate and analyse the commonness and individuality of driving behaviour characteristics with the longitudinal driving behaviour data collected by the instrumented vehicle. To handle the large set of data and discover the valuable knowledge, the data mining techniques including ensemble clustering method based on the kernel fuzzy C-means algorithm and the modified latent Dirichlet allocation model are employed in this study. The ‘aggressive’, ‘cautious’ and ‘moderate’ driving states are discovered and the underlying quantified structure is built for the driving style analysis.

The study presents findings regarding drivers’ patterns of use, attitude towards, and reported effects of access to mature nomadic navigation support systems. Three different systems were tested by 582 drivers in four-field operational tests for a period of six months. A majority of the participants used the support system for trips where the route/destination was unfamiliar but there were also other use scenarios. The main benefits entailed convenience and comfort. Reported effects involved increased possibilities to choose the route according to preferences; a decrease in the time it took to reach destinations and in the distance covered to reach the destination. One in four reported a decrease in fuel consumption attributed an increased compliance with speed limits and/or that driving around and searching for the correct route to reach the desired destination could be avoided. A majority reported ‘no change’ regarding the number of journeys made by car. Reported effects (whether increases or decreases) were however smaller than expected before the trial.

This study proposes an efficient anomalous behaviour detection framework using trajectory analysis. Such framework includes the trajectory pattern learning module and the online abnormal detection module. In the pattern learning module, a coarse-to-fine clustering strategy is utilised. Vehicle trajectories are coarsely grouped into coherent clusters according to the main flow direction (MFD) vectors followed by a three-stage filtering algorithm. Then a robust K-means clustering algorithm is used in each coarse cluster to get fine classification by which the outliers are distinguished. Finally, the hidden Markov model (HMM) is used to establish the path pattern within each cluster. In the online detection module, the new vehicle trajectory is compared against all the MFD distributions and the HMMs so that the coherence with common motion patterns can be evaluated. Besides that, a real-time abnormal detection method is proposed. The abnormal behaviour can be detected when happening. Experimental results illustrate that the detection rate of the proposed algorithm is close to the state-of-the-art abnormal event detection systems. In addition, the proposed system provides the lowest false detection rate among selected methods. It is suitable for intelligent surveillance applications.

The congestion caused by a special type of incident that is different from a normal incident, namely the flooding incident under grade separation bridges, has shown to be a serious issue in Beijing because of several horrible recent experiences. To investigate the characteristics of the congestion, this study strives to develop a floating car data (FCD)-based method for detecting the flooding incident under grade separation bridges. The study first examines the applicability of using an improved cumulative sum (CUSUM) method. However, it is found that the improved CUSUM method does not function properly when all lanes are blocked by the flooding under bridges. Then, the study proposes an analytical method by analysing characteristics of FCD. Three decision parameters, sample losing rate, speed and accumulated discrepancy, are proposed, which play a synergistic effect in the detection. It is shown from case studies that the proposed method performs satisfactorily for detecting flooding incidents under grade separation bridges. The proposed method can be used to further investigate the congestion spreading regularities to develop quick and real-time response process to mitigating the congestion triggered by the flooding.

The use of advanced driver assistance systems is becoming increasingly common in road-going vehicles. One application of these driver assistance systems is in the automated detection of vulnerable road users, such as pedestrians, using automotive far-infrared imagery. Detection of pedestrians in infrared imagery can be quite difficult because of a number of factors such as the environment, pedestrian behaviour and also the physical limitations of currently available infrared sensors. This study presents a comprehensive review of the literature currently available in the area of pedestrian detection techniques in automotive infrared imagery. The challenges associated with automated detection of pedestrians in the automotive domain are first discussed. An overview of the general structure of pedestrian detection algorithms is then presented, followed by an in-depth analysis of existing literature in the area. Some proposals for future research in the area based on the methods described in this study are then offered.