Bayesian vehicle class recognition using 3-d probe

A new approach is presented to vehicle-class recognition from video clips. Two new concepts introduced are: probes consisting of local 3-d curve-groups which when projected into video frames are features for recognizing vehicle classes in video clips; and Bayesian recognition based on class probability densities for groups of 3-d distances between pairs of 3-d probes. A full Bayesian recognizer is realized via Monte Carlo simulation method. Also, a sub-optimal but robust camera calibration method is employed and tested extensively.     

Estimation of soot oxidation rate in DPF under carbon and non-carbon based particulate matter accumulated condition

By high particulate matter(PM) reduction performance, diesel particulate filter(DPF) is applied to almost all of modern HSDI diesel engine. PM emitted from diesel engine is consist of carbon based and non-carbon based material. Representative carbon based PM is soot. Non-carbon based PM is produced by wear of engine and exhaust component, combustion of lubrication oil and sulphur in fuel. Accumulation of non-carbon based PM affects pressure difference of DPF and thus accuracy of soot mass estimation in DPF can be lowered during normal and regeneration condition when the pressure difference caused by non-carbon based PM is not recognized correctly. Also unevenly accumulated PM inside of DPF can produce locally different exhaust gas temperature and thus it can lower accuracy of soot mass estimation during regeneration. This study focuses on estimation of soot oxidation rate not by conventional pressure difference but by exhaust gas analysis at up and downstream of DPF. Results, strong correlations between CO₂ -fuel mass ratio and soot oxidation was observed.     

驾驶模拟在公路安全设计与运行中运用的研究

随着日益增长的交通需求,为了能够获得最高性能的公路交通系统,驾驶任务变得越来越复杂,公路设计和建设中考虑人的因素也变得更为重要.能否成功引进新的驾驶支持体系、动态的交通管理系统或者复杂的道路设计,主要依赖于人们是否能够以及有多大意愿去适应这种发展.为此需要了解人们在复杂的动态驾驶任务环境中将会做出怎样的反应.该文在简单地介绍完TNO驾驶模拟器设备后,又介绍了有关利用先进的人机互动驾驶模拟器研究人的驾驶行为来进行公路设计和建设.文中通过几个公路设计和交通运输行为研究的例子,说明了驾驶模拟器研究对安全有效地进行公路设计与运行运用具有特殊的价值.     

Factors Affecting Disaster Preparedness,Response, and Recovery Using the Community Capitals Framework

Abstract

Disaster research often focuses on how and why communities are affected by a discrete extreme event. We used the community capitals framework to understand how community characteristics influence their preparedness, response to, and recovery from successive or multiple disasters using the 1964 Good Friday Earthquake and the 1989 Exxon Valdez Oil Spill as case studies. This study assesses community response to these disasters by reviewing published literature on impacts to create profiles for six communities and by identifying community capitals before and during these disasters, and throughout the long-term recovery. While the presence of rich natural capitals commonly contributed resources to pre-disaster planning and long-term recovery, restriction of resource access immediately following the disasters was detrimental to many communities. Communities with strong political, social, and financial capitals tended to fare better immediately following disasters, enabling longer-term processes of transformation or recovery. However, in some communities the oil spill undermined these capitals more than the earthquake and resulting tsunami. In understanding how use and reliance on community capitals can lead to varied recovery success from different kinds of disasters, these findings can help coastal managers and planners prepare for future disasters.     

Surveillance and Poaching on Inshore Reefs of the Great Barrier Reef Marine Park

The Great Barrier Reef (GBR), Australia, is managed under the GBR Marine Park Act (1975) and is seen as a shining example of marine resource management. The principle tool of management is zoning for multiple use. We examined surveillance and illegal fishing around two inshore islands (Magnetic and Orpheus) of the GBR Marine Park in 2000/2001. Both islands are near Townsville, the largest city adjacent to the GBR. Surveillance effort was low, with vessels present on only 16% of days of the year. Measurable but low levels of illegal recreational fishing occurred within no-take zones. Levels decreased with increasing surveillance effort. Thus zoning was not completely successful in protecting fish targeted by fisheries, even within the most highly enforced sections of the Park. The expansion of no-take zones in 2004 from 4.6% to 33.4% of the area of the 358,000 km2 Park represents a considerable challenge for future surveillance and enforcement.     

A Framework for Developing Driving Cycles with On‐Road Driving Data

Abstract

This paper attempts to propose a framework on driving cycle development based on a thorough review of 101 transient driving cycles. A comparison of the driving cycles highlighted that Asian driving is the slowest but most aggressive while European driving is the fastest and smoothest. Further review of the cycle development methodologies identified three major elements for developing a driving cycle; test route selection, data collection and cycle construction methods. A framework was eventually proposed based on these findings and recommendations from this review. First, traffic activity patterns and quantitative statistics should be considered in determining the test routes. Speed data can be collected by using chase car method, on‐board measurement techniques or their hybrid. As for the construction of driving cycle, the matching approach has been more commonly used. It is recommended that the tendency of zero change in acceleration, which has been commonly ignored in the literature, and the application of succession probability at second‐by‐second level should be further explored. A fifth mode, creeping, is also recommended for modal analysis for characterizing urban congested driving conditions.     

Establishing speed–flow–density relationships for exclusive motorcycle lanes

Abstract

The motorcycle is a popular mode of transport in Malaysia and developing Asian countries, but its significant representation in the traffic mix results in high rates of motorcycle accidents. As a result, the Malaysian Government decided to segregate motorcycle traffic along its new federal roads as an engineering approach to reduce accidents. However, traffic engineers needed to know the maximum traffic a motorcycle lane could accommodate. Despite substantial literature related to speed–flow–density relationships and capacities of various transport facilities, there is a knowledge gap regarding motorcycle lanes. This paper establishes motorcycle speed–flow–density relationships and capacities of exclusive motorcycle lanes in Malaysia. Observations of motorcycle flows and speeds were conducted along existing and experimental motorcycle lanes. Motorcycle speed–density data were aggregated and plotted for two types of observable motorcycle riding behaviour patterns that were influenced by the widths of a motorcycle lane: the headway pattern (lane width ≤ 1.7 m) and the space pattern (lane width > 1.7 m). For both riding patterns, regression analysis of motorcycle speed–density data best fits the logarithmic model and consequently the motorcycle flow–density and speed–flow models are derived. Motorcycle lane capacities for headway and space riding patterns are estimated as 3300 mc/hr/lane and 2200 mc/hr/m, respectively.     

Modelling trip distribution with fuzzy and genetic fuzzy systems

This paper explores the potential capabilities of fuzzy and genetic fuzzy system approaches in urban trip distribution modelling with some new features. First, a simple fuzzy rule-based system (FRBS) and a novel genetic fuzzy rule-based system [GFRBS: a fuzzy system improved by a knowledge base learning process with genetic algorithms (GAs)] are designed to model intra-city passenger flows for Istanbul. Subsequently, their accuracy, applicability and generalizability characteristics are evaluated against the well-known gravity- and neural network (NN)-based trip distribution models. The overall results show that: traditional doubly constrained gravity models are still simple and efficient; NNs may not show expected performance when they are forced to satisfy trip constraints; simply-designed FRBSs, learning from observations and expertise, are both efficient and interpretable even if the data are large and noisy; and use of GAs in fuzzy rule-based learning considerably increases modelling performance, although it brings additional computation cost.     

Accident severity analysis using ordered probit model

To reduce injuries in road crashes, better understanding is needed between the relationship of injury severity and risk factors. This study seeks to identify the contributing factors affecting crash severity with broad considerations of driver characteristics, roadway features, vehicle types, pedestrian characteristics and crash characteristics using an ordered probit model. It also explores how the interaction of these factors will affect accident severity risk. Three types of accidents were investigated: two-vehicle crashes, single vehicle crashes and pedestrian accidents. The reported crash data in Singapore from 1992 to 2001 were used to illustrate the process of parameter estimation. Several factors such as vehicle type, road type, collision type, location type, pedestrian age, time of day of accident occurrence were found to be significantly associated with injury severity. It was also found that injury severity decreases over time for the three types of accident investigated.