Concurrent multi-scale modeling of civil infrastructures for analyses on structural deteriorating—Part II : Model updating and verification

This paper is a continuation of the paper titled “Concurrent multi-scale modeling of civil infrastructure for analyses on structural deteriorating—Part I: Modeling methodology and strategy” with the emphasis on model updating and verification for the developed concurrent multi-scale model. The sensitivity-based parameter updating method was applied and some important issues such as selection of reference data and model parameters, and model updating procedures on the multi-scale model were investigated based on the sensitivity analysis of the selected model parameters. The experimental modal data as well as static response in terms of component nominal stresses and hot-spot stresses at the concerned locations were used for dynamic response- and static response-oriented model updating, respectively. The updated multi-scale model was further verified to act as the baseline model which is assumed to be finite-element model closest to the real situation of the structure available for the subsequent arbitrary numerical simulation. The comparison of dynamic and static responses between the calculated results by the final model and measured data indicated the updating and verification methods applied in this paper are reliable and accurate for the multi-scale model of frame-like structure. The general procedures of multi-scale model updating and verification were finally proposed for nonlinear physical-based modeling of large civil infrastructure, and it was applied to the model verification of a long-span bridge as an actual engineering practice of the proposed procedures.

Fusion of in-vehicle sensor data to develop Intelligent Driver Training System (IDTS)

The over represented number of novice drivers involved in crashes is alarming. Driver training is one of the interventions aimed at mitigating the number of crashes that involve young drivers. To our knowledge, Advanced Driver Assistance Systems (ADAS) have never been comprehensively used in designing an intelligent driver training system. Currently, there is a need to develop and evaluate ADAS that could assess driving competencies. The aim is to develop an unsupervised system called Intelligent Driver Training System (IDTS) that analyzes crash risks in a given driving situation. In order to design a comprehensive IDTS, data is collected from the Driver, Vehicle and Environment (DVE), synchronized and analyzed. The first implementation phase of this intelligent driver training system deals with synchronizing multiple variables acquired from DVE. RTMaps is used to collect and synchronize data like GPS, vehicle dynamics and driver head movement. After the data synchronization, maneuvers are segmented out as right turn, left turn and overtake. Each maneuver is composed of several individual tasks that are necessary to be performed in a sequential manner. This paper focuses on turn maneuvers. Some of the tasks required in the analysis of ‘turn’ maneuver are: detect the start and end of the turn, detect the indicator status change, check if the indicator was turned on within a safe distance and check the lane keeping during the turn maneuver. This paper proposes a fusion and analysis of heterogeneous data, mainly involved in driving, to determine the risk factor of particular maneuvers within the drive. It also explains the segmentation and risk analysis of the turn maneuver in a drive.

Combination strategies for a factor analysis phone-conditioned speaker verification system

This work aims to take advantage of recent developments in joint factor analysis (JFA) in the context of a phonetically conditioned GMM speaker verification system. Previous work has shown performance advantages through phonetic conditioning, but this has not been shown to date with the JFA framework. Our focus is particularly on strategies for combining the phone-conditioned systems. We show that the classic fusion of the scores is suboptimal when using multiple GMM systems. We investigate several combination strategies in the model space, and demonstrate improvement over score-level combination as well as over a non-phonetic baseline system. This work was conducted during the 2008 CLSP Workshop at Johns Hopkins University.

Articulating cooperative systems and driver behaviour theories

There has been increased research interest in Co-operative Vehicle Infrastructure Systems (CVIS) from the eld of Intelligent Transport Systems (ITS). However most of the research have focused on the engineering aspects and overlooked their relevance to the drivers' behaviour. This paper argues that the priority for cooperative systems is the need to improve drivers decision making and reduce drivers' crash risk exposure to improve road safety. Therefore any engineering solutions need to be considered in conjuction with traffic psychology theories on driver behaviour. This paper explores the advantages and limitations of existing systems and emphasizes various theoretical issues that arise in articulating cooperative systems' capabilities and drivers' behaviour.

Predicting driver's hypovigilance on monotonous roads: literature review.

Drivers' ability to react to unpredictable events deteriorates when exposed to highly predictable and uneventful driving tasks. Particularly, highway design reduces the driving task mainly to a lane-keeping one. It contributes to hypovigilance and road crashes as drivers are often not aware that their driving behaviour is impaired. Monotony increases fatigue, however, the fatigue community has mainly focused on endogenous factors leading to fatigue such as sleep deprivation. This paper focuses on the exogenous factor monotony which contributes to hypovigilance. Objective measurements of the effects of monotonous driving conditions on the driver and the vehicle's dynamics is systematically reviewed with the aim of justifying the relevance of the need for a mathematical framework that could predict hypovigilance in real-time. Although electroencephalography (EEG) is one of the most reliable measures of vigilance, it is obtrusive. This suggests to predict from observable variables the time when the driver is hypovigilant. Outlined is a vision for future research in the modelling of driver vigilance decrement due to monotonous driving conditions. A mathematical model for predicting drivers’ hypovigilance using information like lane positioning, steering wheel movements and eye blinks is provided. Such a modelling of driver vigilance should enable the future development of an in-vehicle device that detects driver hypovigilance in advance, thus offering the potential to enhance road safety and prevent road crashes.

Correlating eye gaze direction, depth and vehicle information on an interactive map for driver training

The over represented number of novice drivers involved in crashes is alarming. Driver training is one of the interventions aimed at mitigating the number of crashes that involve young drivers. Experienced drivers have better hazard perception ability compared to inexperienced drivers. Eye gaze patterns have been found to be an indicator of the driver's competency level. The aim of this paper is to develop an in-vehicle system which correlates information about the driver's gaze and vehicle dynamics, which is then used to assist driver trainers in assessing driving competency. This system allows visualization of the complete driving manoeuvre data on interactive maps. It uses an eye tracker and perspective projection algorithms to compute the depth of gaze and plots it on Google maps. This interactive map also features the trajectory of the vehicle and turn indicator usage. This system allows efficient and user friendly analysis of the driving task. It can be used by driver trainers and trainees to understand objectively the risks encountered during driving manoeuvres. This paper presents a prototype that plots the driver's eye gaze depth and direction on an interactive map along with the vehicle dynamics information. This prototype will be used in future to study the difference in gaze patterns in novice and experienced drivers prior to a certain manoeuvre.

In-vehicle driver distractions : characteristics underlying drivers' risk perceptions

Driver distraction continues to receive considerable research interest but the drivers‟ perspective is less well documented. The current research focussed on identifying features that are salient to drivers in their risk perception judgements for 19 in-vehicle distractions. Both technological (e.g. mobile phones) and non technological (e.g. eating) distractions were considered. Analysis identified that males and females were rating 7 of the 19 distractions differently. The current paper presents the data for the female participants (n = 84). Multidimensional scaling analysis identified three main dimensions contributing to female drivers‟ risk perception judgements. Qualitative characteristics such as the level of exposure to a distraction were identified as significant contributors to drivers‟ risk perception as well as features inherent in the distractions such as distractions being related to communication. This exploratory work contributes to better understanding female drivers‟ perceptions of risk associated with in-vehicle distractions. Understanding the drivers‟ perspective can help guide the development of road safety messages and ultimately improve the impact of such messages.

Barriers and facilitators of a designated driver program

The general aim of designated driver programs is to reduce the level of drink driving by encouraging potential drivers to travel with a driver who has abstained from (or at least limited) consuming alcohol. Designated driver programs are quite widespread around the world, however a limited number have been rigorously evaluated. This paper reports the qualitative results from an evaluation of a designated driver program known as ‘Skipper’, in a provincial city in Queensland. Focus groups were conducted with 108 individuals from the intervention area. These focus groups aimed to assess the barriers and facilitators to the programs’ effectiveness by obtaining information about the patrons’ views on various aspects of the program, as well as designated driver and travelling after drinking more generally. A brief questionnaire was also given to participants in order to present responses in terms of the participants’ characteristics. Results suggest general support for the designated driver concept and the ‘Skipper’ program specifically. Facilitating factors reported by participants included the media coverage highlighting the risks associated with drink driving and the social acceptability of choosing not to drink. However, there was also some suggestion that the impact of the program was mainly to encourage those who already engage in designated driver behaviour to continue doing so, rather than encouraging the uptake of the behaviour among potential new users. Some of the suggested barriers to this kind of behaviour change include: social pressure to drink; alcohol dependency; and a failure to plan ahead.

Enhancing road safety for young drivers : how Graduate Driver Licensing initiatives can complement "anti-hooning" legislation

Street racing and associated (hooning) behaviours have attracted increasing concern in recent years. While New Zealand and all Australian jurisdictions have introduced “antihooning” legislation and allocated significant police resources to managing the problem, there is limited evidence of the road safety implications of hooning. However, international and Australian data suggests that drivers charged with a hooning offence tend to be young males who are accompanied by one or more peers, and hooning-related crashes tend to occur at night. In this regard, there is considerable evidence that drivers under the age of 25 are over-represented in crash statistics, and are particularly vulnerable soon after obtaining a Provisional licence, when driving at night, and when carrying peer-aged passengers. The similarity between the nature of hooning offenders, offences and crashes, and road safety risks for young drivers in general, suggests that hooning is an issue that may be viewed as part of the broader young driver problem. Many jurisdictions have recently implemented a range of evidence-based strategies to address young driver road safety, and this paper will present Queensland crash and offence data to highlight the potential benefit of Graduated Driver Licensing initiatives, such as night driving restrictions and peer-aged passenger restrictions, to related road safety issues, including hooning. An understanding of potential flow-on effects is important for evaluations of anti-hooning legislation and Graduated Driver Licensing programs, and may have implications for future law enforcement resource allocation and policy development.

Improved GMM-based speaker verification using SVM-driven impostor dataset selection

The problem of impostor dataset selection for GMM-based speaker verification is addressed through the recently proposed data-driven background dataset refinement technique. The SVM-based refinement technique selects from a candidate impostor dataset those examples that are most frequently selected as support vectors when training a set of SVMs on a development corpus. This study demonstrates the versatility of dataset refinement in the task of selecting suitable impostor datasets for use in GMM-based speaker verification. The use of refined Z- and T-norm datasets provided performance gains of 15% in EER in the NIST 2006 SRE over the use of heuristically selected datasets. The refined datasets were shown to generalise well to the unseen data of the NIST 2008 SRE.

Data-driven impostor selection for T-norm score normalisation and the background dataset in SVM-based speaker verification

A data-driven background dataset refinement technique was recently proposed for SVM based speaker verification. This method selects a refined SVM background dataset from a set of candidate impostor examples after individually ranking examples by their relevance. This paper extends this technique to the refinement of the T-norm dataset for SVM-based speaker verification. The independent refinement of the background and T-norm datasets provides a means of investigating the sensitivity of SVM-based speaker verification performance to the selection of each of these datasets. Using refined datasets provided improvements of 13% in min. DCF and 9% in EER over the full set of impostor examples on the 2006 SRE corpus with the majority of these gains due to refinement of the T-norm dataset. Similar trends were observed for the unseen data of the NIST 2008 SRE.

Within-session variability modelling for factor analysis speaker verification

This work presents an extended Joint Factor Analysis model including explicit modelling of unwanted within-session variability. The goals of the proposed extended JFA model are to improve verification performance with short utterances by compensating for the effects of limited or imbalanced phonetic coverage, and to produce a flexible JFA model that is effective over a wide range of utterance lengths without adjusting model parameters such as retraining session subspaces. Experimental results on the 2006 NIST SRE corpus demonstrate the flexibility of the proposed model by providing competitive results over a wide range of utterance lengths without retraining and also yielding modest improvements in a number of conditions over current state-of-the-art.

Interaction between visual status, driver age and distracters on daytime driving performance

This study investigated the effects of visual status, driver age and the presence of secondary distracter tasks on driving performance. Twenty young (M = 26.8 years) and 19 old (M = 70.2 years) participants drove around a closed-road circuit under three visual (normal, simulated cataracts, blur) and three distracter conditions (none, visual, auditory). Simulated visual impairment, increased driver age and the presence of a distracter task detrimentally affected all measures of driving performance except gap judgments and lane keeping. Significant interaction effects were evident between visual status, age and distracters; simulated cataracts had the most negative impact on performance in the presence of visual distracters and a more negative impact for older drivers. The implications of these findings for driving behaviour and acquisition of driving-related information for people with common visual impairments are discussed