Development of probabilistic pedestrian fatality model for characterizing pedestrian-vehicle collisions

Pedestrian-related accidents are considered to be the most serious of traffic accidents due to the associated high fatality rates. In Korea, pedestrian fatalities accounted for approximately 40% of all traffic-related fatalities in 2004. Significant efforts have been made to develop effective countermeasures for pedestrian-vehicle collisions. A basis for devising such countermeasures is to understand the characteristics of pedestrian-vehicle collisions. This study develops a pedestrian fatality model capable of predicting the probability of fatality in pedestrian-vehicle collisions. Binary logistic regression and a probabilistic neural network (PNN) are employed to estimate the probability of pedestrian fatality. Pedestrian age, vehicle type and collision speed are used as independent variables of the fatality model. The models developed herein are valuable tools that can be used to direct safety policies and technologies associated with pedestrian safety.     

A two-stage-training support vector machine approach to predicting unintentional vehicle lane departure

Advanced driver assistance systems, such as unintentional lane departure warning systems, have recently drawn much attention and efforts. In this study, we explored utilizing the nonlinear binary support vector machine (SVM) technique to predict unintentional lane departure, which is innovative, as the SVM methodology has not previously been attempted for this purpose in the literature. Furthermore, we developed a two-stage training scheme to improve SVM's prediction performance in terms of minimization of the number of false positive prediction errors. Experiment data generated by VIRTTEX, a hydraulically powered, 6-degrees-of-freedom moving base driving simulator at Ford Motor Company, were used. All the vehicle variables were sampled at 50 Hz and there were 16 drowsy drivers (about 3 hours of driving per subject) and six control drivers (approximately 20 minutes f driving each). In total, 3,508 unintentional lane departures occurred for the drowsy drivers and 23 for the control drivers. Our study involving these 22 drivers with a total of more than 7.5 million prediction decisions demonstrates that (a) excellent SVM prediction performance, measured by numbers of false positives (i.e., falsely predicted lane departures) and false negatives (i.e., lane departures failed to be predicted), was achieved when the prediction horizon was 0.6 seconds or less, (b) lateral position and lateral velocity worked the best as SVM input variables among the nine variable sets that we explored, and (c) the radial basis function performed the best as the SVM kernel function.     

Gain scheduled linear quadratic tracking system tuned optimally by covariance matrix adaption evolutionary strategy for automotive engine coldstart control

In this paper, a gain scheduled linear quadratic tracking system (LQTS) tuned optimally by an evolutionary strategy (ES) is devised to reduce the total tailpipe hydrocarbon (HC) emissions of an automotive engine over the coldstart period. As the engine’s behavior during coldstart operations is nonlinear, the system dynamics is clearly analyzed and represented by a number of separate linear models generated based on a coldstart model verified by experimental data. An independent LQTS is then implemented for each of these linear models. In this way, several control laws are created, and the corresponding gains are calculated for each of the independent control laws. ES is then used to tune the adjustable parameters of LQTSs to calculate the control inputs, namely air/fuel ratio (AFR) and spark timing (Δ), such that the resulting exhaust gas temperature (T _exh) and engine-out HC emissions (HC _raw) be close to a set of optimum profiles. This enables the controller reduce the cumulative tailpipe hydrocarbon emissions (HC _cum) to the highest possible extent. To demonstrate the acceptable performance of the proposed controller, an optimal controller derived from the Pontryagin’s minimum principle (PMP) is also taken into account. Based on the results of the conducted comparative study, it is shown that the proposed control technique has a very good performance, and also, can be easily used for real-time applications, as it consumes a remarkably trivial computational time for calculating the controlling commands.     

Brazil’s new vehicle front impact safety standards — ABNT NBR 15300-1 & ABNT NBR 15300-2 or ABNT NBR 15300-3: Is the manufacturerer choice the safest for the passenger?

To increase car passenger safety, the Brazilian National Traffic Council (CONTRAN) released Resolution 221, which defines the maximum passenger and driver biomechanical criteria in the event of a vehicle frontal impact. The vehicle maximum allowed biomechanical injury criteria will be enforced from January 2012 for new vehicles and in January 2014 for vehicles in production before January 2014. To standardize the test method to measure the driver and front passenger injury values in a frontal crash, Resolution 221 states that the tests must be performed according to the ABNT NBR 15300-1 standard, followed by the ABNT NBR 15300-2 standard or the ABNT NBR 15300-3 standard. The use of ABNT NBR 15300-2 or ABNT NBR 15300-3 standards is a free choice for the manufacturer of the vehicle. The ABNT NBR 15300-1 + 15300-2 test is similar to the FMVSS 208 standard in the United States in terms of its vehicle frontal impact test perpendicular to a rigid barrier with the use of seat belts by male model dummies. The test according to ABNT NBR 15300-1 + 15300-3 follows the European ECE R94 and 96/79/EC standards. However, ABNT NBR 15300-2 focuses on occupant protection during vehicle deceleration rather than occupant protection during vehicle deformation in a crash test. ABNT NBR 15300-3 tests occupant protection during vehicle deformation more than it tests occupant protection during vehicle deceleration. Therefore, this paper aims to show the types of test results produced by the ABNT NBR 15300-2 and ABNT NBR 15300-3 standards and their differences concerning occupant protection verification and discuss the manufacturer??s freedom of choice.     

Fuzzy expert system for controlling swamp terrain intelligent air-cushion tracked vehicle

A fuzzy expert system was used in this study to control an intelligent air-cushion tracked vehicle (IACTV) as it operated in a swamp peat terrain. The system was effective in controlling the intelligent air-cushion vehicle while measuring the vehicle traction (TE), motion resistance (MR), power consumption (PC), cushion clearance height (CCH) and cushion pressure (CP). An ultrasonic displacement sensor, pull-in solenoid electromagnetic switch, pressure-control sensor, microcontroller, and battery pH sensor were incorporated into the fuzzy expert system (FES) to experimentally determine the TE, MR, PC, CCH, and CP. In this study, we provide an illustration of how an FES might play an important role in the prediction of the power consumption of the vehicle’s intelligent air-cushion system. The mean relative error in the actual and predicted values from the FES model with respect to tractive effort, total motion resistance and total power consumption were found to be 5.58 %, 6.78 % and 10.63 %, respectively. For all parameters, the relative error in the predicted values was found to be less than the acceptable limit (10%), except for the total power consumption. Furthermore, the goodness of fit of the predicted values was found to be close to 1.0 as expected and, hence, indicates the good performance of the developed system.     

Transport demand model estimation from traffic counts

Many problems in transport planning and management tasks require an origindestination (O-D) matrix to represent the travel pattern. However, O-D matrices obtained through a large scale survey such as home or roadside interviews, tend to be costly, labour intensive and time disruptive to trip makers. Therefore, the use of low cost and easily available data is particularly attractive.The need of low-cost methods to estimate current and future O-D matrices is even more valuable in developing countries because of the rapid changes in population, economic activity and land use. Models of transport demand have been used for many years to synthesize O-D matrices in study areas. A typical example of this is the gravity model; its functional form, plus the appropriate values for the parameters involved, is employed to produce acceptable matrices representing trip making behaviour for many trip purposes and time periods.The work reported in this paper has combined the advantages of acceptable travel demand models with the low cost and availability of traffic counts. Three types of demand models have been used: gravity (GR), opportunity (OP) and gravity-opportunity (GO) models. Three estimation methods have been developed to calibrate these models from traffic counts, namely: non-linear-least-squares (NLLS), weighted-non-linear-least-squares (WNLLS) and maximumlikelihood (ML).The 1978 Ripon (urban vehicle movement) survey was used to test these methods. They were found to perform satisfactorily since each calibrated model reproduced the observed O-D matrix fairly closely. The tests were carried out using two assignment techniques, all-or-nothing and the stochastic method due to Burrell, in determining the routes taken through the network.requests for offprints     

Analysis of Metropolitan Highway Capacity and the growth in vehicle miles of travel

A number of recent studies have examined the hypothesis of induced travel in an attempt to quantify the phenomenon (Hansen & Huang 1997; Noland, forthcoming). No study has yet attempted to adjust for potential simultaneity bias in the results. This study addresses this issue by the use of an instrumental variable (two stage least squares) approach. Metropolitan level data compiled by the Texas Transportation Institute for their annual congestion report is used in the analysis and urbanized land area is used as an instrument for lane miles of capacity. While this is not an ideal instrument, results still suggest a strong causal relationship but probably that most previous work has had an upward bias in the coefficient estimates. The effect of lane mile additions on VMT growth is forecast and found to account for about 15% of annual VMT growth with substantial variation between metropolitan areas. This effect appears to be closely correlated with percent growth in lane miles, suggesting that rapidly growing areas can attribute a greater share of their VMT growth to growth in lane miles.     

A spatial decision support system for retail plan generation and impact assessment

Current geographic information systems typically offer limited analytical capabilities and lack the flexibility to support spatial decision making effectively. Spatial decision support systems aim to fill this gap. Following this approach, this paper describes an operational system for integrated land-use and transportation planning called Location Planner. The system integrates a wide variety of spatial models in a flexible and easy-to-use problem solving environment. Users are able to construct a model out of available components and use the model for impact analysis and optimization. Thus, in contrast to existing spatial decision support systems, the proposed system allows users to address a wide range of problems. The paper describes the architecture of the system and an illustrative application. Furthermore, the potentials of the system for land-use and transportation planning are discussed.     

Analysis of activity duration using the Puget sound transportation panel

In this paper, we extend previous analyses of temporal effects by examining the factors associated with activity generation and duration over four waves of the Puget sound transportation panel survey (PSTPS). A Cox proportional hazards model was specified for each of five activities: visiting, appointment, free time, personal business, and shopping. For each activity, the duration times are modeled with an emphasis on examining important higher order interactions. The results suggest that activity durations have changed significantly over the survey period. Many of the differences in activity durations over time were significant, and often associated with increasing numbers of children in the household and higher order interactions between sex and the sequencing of activities.