A similarity-based neuro-fuzzy modeling for driving behavior recognition applying fusion of smartphone sensors

Drivers’ behavior evaluation is one of the most important problems in intelligent transportation systems and driver assistant systems. It has a great influence on driving safety and fuel consumption. One of the challenges in this regard is the modeling perspective to treat with uncertainty in judgments about driving behaviors. Really, assessing a single maneuver with a rigid threshold leads to a weak judgment for driving evaluation. To fill this gap, a novel neuro-fuzzy system is proposed to classify the driving behaviors based on their similarities to fuzzy patterns when all of the various maneuvers are stated with some fuzzy numbers. These patterns are also fuzzy numbers and they are extracted from statistical analysis on the smartphone sensors data. Our driving evaluation system consists of three processes. Firstly, it detects the type of all of the maneuvers through the driving period, by using a multi-layer perceptron neural network. Secondly, it extracts a new feature based on the acceleration and assigns three fuzzy numbers to driver’s lane change, turn and U-turn maneuvers. Thirdly, it determines the similarity between these three fuzzy numbers and the fuzzy patterns to evaluate the safe and the aggressive driving scores. To validate this model, Driver’s Angry Score (DAS) questionnaires are used. Results show that the fusion of Inertial Measurement Unit (IMU) sensors of smartphones is enough for the proposed driving evaluation system. Accuracy of this system is 87% without using GPS and GIS data and this system is independent of smartphones and vehicles types.     

Factors contributing to the severity of intersection crashes

Road crashes are a leading cause of death and serious injuries both developed and developing countries. Intersections are recognized as being among the most hazardous locations on the roads. Although crashes at intersections form about 35 % of the reported accidents account for about 32% of traffic‐related serious injuries and fatalities in Singapore, there is no known study that examines the factors contributing to the severity of these crashes. In this study, the ordinal probit model was applied to crash data from 1992 to 2002 to investigate the role a variety of factors play in determining the severity of intersection crashes. Our study shows that vehicle type, road type, collision type, driver's characteristics and time of day are important determinants of the severity of crashes at intersections in Singapore.     

Optimal fueling strategies for locomotive fleets in railroad networks

Railroad companies spend billions of dollars each year to purchase fuel for thousands of locomotives across the railroad network. Each fuel station charges a site-dependent fuel price, and the railroad companies must pay an additional flat contracting fee in order to use it. This paper presents a linear mixed-integer mathematical model that integrates not only fuel station location decisions but also locomotive fueling schedule decisions. The proposed model helps railroads decide which fuel stations to contract, and how each locomotive should purchase fuel along its predetermined shipment path, such that no locomotive runs out of fuel while the summation of fuel purchasing costs, shipment delay costs (due to fueling), and contracting charges is minimized. A Lagrangian relaxation framework is proposed to decompose the problem into fueling schedule and facility location selection sub-problems. A network shortest path formulation of the fueling schedule sub-problem is developed to obtain an exact optimal solution to the fueling schedule sub-problem. The proposed framework is applied to a large-scale empirical case and is shown to effectively reduce system costs.     

A numerical investigation into strength and deformation characteristics of preloaded tubular members under lateral impact loads

This paper presents the results of a numerical investigation into the structural behaviour of preloaded tubular members under lateral impact loads by means of finite element method. The lateral load represents a statically modelled impact from collision between tubular member and a solid rectangular indenter. Three different kinds of end conditions have been applied to the model and the effects of boundary conditions are investigated. Also, the effect of preloading on the buckling strength as well as the ultimate strength for laterally impacted tubes is assessed and it will be shown that preloading and position of applying force directly affect these strengths. In other words, by increasing in the amount of preloading, ultimate strength reduces and member tends to collapse under lower amounts of loads. The influence of the position of applying lateral load has also been addressed and relevant results will be discussed. In order to verify the performance of numerical model, the results have been examined against an available experimental test.     

Estimating intersection O-D matrices from observed time-series of traffic counts: The zellner estimator

The problem of estimating intersection O-D matrices from input and output time-series of traffic counts is considered in this paper. Because of possible existence of significant correlation between the error terms across structural equations forming the O-D matrices, the seemingly unrelated estimator (Zellner estimator) was suggested. Estimation results showed evidence of strong correlation between error terms across-equations. Generally, the Zellner estimator produced more efficient estimates than did the ordinary least-squares estimator. Furthermore, the Zellner estimator satisfied all constraints and reproduced turning movements comparable to the actual ones.     

A Fuzzy Logic Direct Yaw-Moment Control System for All-Wheel-Drive Electric Vehicles

Summary In-wheel-motors are revolutionary new electric drive systems that can be housed in vehicle wheel assemblies. Such E-wheels permit packaging flexibility by eliminating the central drive motor and the associated transmission and driveline components, including the transmission, the differential, the universal joints and the drive shaft. Apart from many advantages of such a system, unequalled independent wheel control allows vehicle dynamic improvement to assist the driver in enhancing cornering and straight-line stability on slippery roads and in adverse ground conditions. In this paper a Fuzzy logic driver-assist stability system for all-wheel-drive electric vehicles based on a yaw reference DYC is introduced. The system assists the driver with path correction, thus enhancing cornering and straight-line stability and providing enhanced safety. A feed-forward neural network is employed to generate the required yaw rate reference. The neural net maps the vehicle speed and the steering angle to give the yaw rate reference. The vehicle true speed is estimated using a multi-sensor data fusion method. Data from wheel sensors and an embedded accelerometer are fed into an estimator, where a Fuzzy logic system decides which input is more reliable. The efficiency of the proposed system is approved by conducting a computer simulation. The proposed control system is an effective and easy to implement method to enhance the stability of all-wheel-drive electric vehicles.     

Modeling to Predict Rollover Threat of Tractor-Semitrailers

Summary A predictive model to determine a rollover threat index associated with tractor-semitrailers is proposed. The purpose of this model is to predict the rollover threat sufficiently in advance of the actual event so as to enable the driver to react accordingly. The predictive model is established using simple roll-plane models of the vehicle sprung and unsprung masses in conjunction with online vehicle parameter identification. Using this predictive model, the predicted Load Transfer Ratio (LTR) for the trailer axle can be determined as the rollover threat index. The proposed predictive model and the associated parameter-identification algorithm are verified using a 12-degree-of-freedom vehicle model. It is shown that the identified parameter values are close to the actual ones used in detailed simulation study. Similarly it is shown that the predicted values of the LTR are close to the simulated ones, and hence the proposed approach is potentially suitable as the basis for application in rollover threat assessment.     

Reliability?Based Analysis of a Caisson Breakwater with the Application of Bayesian Inference

Caisson breakwaters are mainly constructed in deep waters to protect an area against waves. These breakwaters are con-ventionally designed based on the concept ...     

A new twist on the gig economy: conducting surveys on Amazon Mechanical Turk

Transportation - There is growing interest in using online outsourcing platforms that are part of the “gig economy” to conduct surveys for academic research. This interest has been...