Search Results

The proposed technique is a tailored deep neural network (DNN) training approach which uses an iterative process to support the learning of DNNs by targeting their specific misclassification and missed detections. The process begins with a DNN that is trained on freely available annotated image data, which we will refer to as the Base model, where a subset of the categories for the classifier are related to the automotive theater. A small set of video capture files taken from drives with test vehicles are selected, (based on the diversity of scenes, frequency of vehicles, incidental lighting, etc.), and the Base model is used to detect/classify images within the video files. A software application developed specifically for this work then allows for the capture of frames from the video set where the DNN has made misclassifications. The corresponding annotation files for these images are subsequently corrected to eliminate mislabels.

Ford Motor Company has recently implemented a Hardware-In-the-Loop (HIL) testing system for a new, highly complex, hybrid electric vehicle (HEV) Electronic Control Unit (ECU). The implementation of this HIL system has been quick and effective, since it is based on proven Commercial-Off-The-Shelf (COTS) automation tools for real-time that allow for a very flexible and intuitive design process. An overview of the HIL system implementation process and the derived development benefits will be shown in this paper. The initial concept for the use of this HIL system was a complete closed-loop vehicle simulation environment for Vehicle System Controller testing, but the paper will show that this concept has evolved to allow for the use of the HIL system for many facets of the design process.

There are several reasons why it can be daunting to apply Six Sigma to product creation. Foremost among them, the functional performance of new technologies is unknown prior to starting a project. Although, Design For Six Sigma (DFSS) was developed to overcome this difficulty, a lack of applicable in-class case studies makes it challenging to train the product creation community. The current paper describes an in-class project which illustrates how Six Sigma is applied to a simulated product creation environment. A toy construction set (TCS) project is used to instruct students how to meet customer expectations without violating cost, packaging volume and design-complexity constraints.

Subjective steering feel tuning and objective verification tests are conducted on vehicle prototypes that are a subset of the total number of buildable combinations of body style, drivetrain and tires. Limited development time, high prototype vehicle cost, and hence limited number of available prototypes are factors that affect the ability to tune and verify all the possible configurations. A new model-based process and a toolset have been developed to enhance the existing steering development process such that steering tuning efficiency and performance robustness can be improved. The innovative method utilizes the existing vehicle dynamics simulation and/or physical test data in conjunction with steering system control models, and provides users with simple interfaces which can be used by either CAE or development engineers to perform virtual tuning of the vehicle steering feel to meet performance targets.

The transportation sector is facing three revolutions: shared mobility, electrification, and autonomous driving. To inform decision making and guide smart transportation system development at the city-level, it is critical to model and evaluate how travelers will behave in these systems. Two key components in such models are (1) individual travel demands with high spatial and temporal resolutions, and (2) travelers’ sociodemographic information and trip purposes. These components impact one’s acceptance of autonomous vehicles, adoption of electric vehicles, and participation in shared mobility. Existing methods of travel demand generation either lack travelers’ demographics and trip purposes, or only generate trips at a zonal level. Higher resolution demand and sociodemographic data can enable analysis of trips’ shareability for car sharing and ride pooling and evaluation of electric vehicles’ charging needs.

The performance of structure-borne road NVH can be cascaded down to three major systems: 1) vehicle body structure, 2) chassis/suspension, 3) tire/wheel. The forces at the body attachment points are controlled by the isolation efficiency of the chassis/suspension system and the excitation at the spindle/knuckle due to the tire/road interaction. The chassis force transmissibility is a metric to quantify the isolation efficiency. This paper presents a new experimental methodology to estimate the chassis force transmissibility from a fully assembled vehicle. For the calculation of the transmissibility, the spindle force/moment estimation and the conventional Noise Path Analysis (NPA) methodologies are utilized. A merit of the methodology provides not only spindle force to body force transmissibility but also spindle moment to body force transmissibility. Hence it enables us to understand the effectiveness of the spindle moments on the body forces.

Today's automotive and electronics technologies are evolving so rapidly that educators and industry are both challenged to re-educate the technological workforce in the new area before they are replaced with yet another generation. In early November 2009 Ford's Product Development senior management formally approved a proposal by the University of Detroit Mercy to transform 125 of Ford's “IC Engine Automotive Engineers” into “Advanced Electric Vehicle Automotive Engineers.” Two months later, the first course of the Advanced Electric Vehicle Program began in Dearborn. UDM's response to Ford's needs (and those of other OEM's and suppliers) was not only at the rate of “academic light speed,” but it involved direct collaboration of Ford's electric vehicle leaders and subject matter experts and the UDM AEV Program faculty.

The forming effects along with strain rate, actual material properties and weld effects have been found to be very critical for accurate prediction of crash responses especially the prediction of local deformation. As a result, crash safety engineers started to consider these factors in crash models to improve the accuracy of CAE prediction and reduce prototype testing. The techniques needed to incorporate forming simulation results, including thickness change, residual stresses and strains, in crash models have been studied extensively and are well known in automotive CAE community. However, a challenge constantly faced by crash safety engineers is the availability of forming simulation results, which are usually supplied by groups conducting forming simulations. The forming simulation results can be obtained by either using incremental codes with actual stamping processes or one-step codes with final product information as a simplified approach.

The world urbanization is growing rapidly, bringing many challenges for people to move in dense metropolitan regions. Public transportation is not able to attend the whole demand, and individual transportation modes are struggling with traffic congestion and stringent regulations to reduce its attractiveness, such as the license plate restriction in São Paulo. On the other hand, enablers like smartphones mass penetration, GPS connected services and shared economy have opened space to a whole new range of possible solutions to improve people perception on urban mobility. This work aims to evaluate the modal choice behavior models and understand the success factor of current mobility solutions in the city of São Paulo. The data available through origin/destination researches will be used to validate the models used in this work.

Thermal modeling of liquid-cooled vehicle traction battery assemblies using Computational Fluid Dynamics (CFD) usually involves large models to accurately resolve small cooling channel details, and intensive computation to simulate drive-cycle transient solutions. This paper proposes a segregated method to divide the system into three parts: the cells, the cold plate and the interface between them. Each of the three parts can be separated and thermally characterized and then combined to predict the overall system thermal behavior for both steady-state and transient operating conditions. The method largely simplifies battery thermal analysis to overcome the limitations of using large 3D CFD models especially for pack level dynamic drive cycle simulations.

Plug in hybrid electric vehicles (PHEVs) have gained interest over last decade due to their increased fuel economy and ability to displace some petroleum fuel with electricity from power grid. Given the complexity of this vehicle powertrain, the energy management plays a key role in providing higher fuel economy. The energy management algorithm on PHEVs performs the same task as a hybrid vehicle energy management but it has more freedom in utilizing the battery energy due to the larger battery capacity and ability to be recharged from the power grid. The state of charge (SOC) profile of the battery during the entire driving trip determines the electric energy usage, thus determining overall fuel consumption.

This work presents a study of crash energy and severity in frontal offset Vehicle-To-Vehicle (VTV) crash tests. The crash energy is analyzed based on analytical formulations and empirical data. Also, the crash severity of different VTV tests is analyzed and compared with the corresponding full frontal rigid barrier test data. In this investigation, the Barrier Equivalent Velocity (BEV) concept is used to calculate the initial impact velocity of frontal offset VTV test modes such that the offset VTV tests are equivalent to full frontal impact tests in terms of crash severity. Linear spring-mass model and collinear impact assumptions are used to develop the mathematical formulation. A scale factor is introduced to account for these assumptions and the calculated initial velocity is adjusted by this scale factor. It is demonstrated that the energies due to lateral and rotational velocity components are very small in the analyzed frontal VTV tests.

Researchers and engineers are utilizing big data analytics to draw further insights into transportation systems. Large amounts of data at the individual vehicle trip level are being collected and stored. The true potential of such data is still to be determined. In this paper, we are presenting a data-driven, novel, and intuitive approach to model driver behaviors using microscopic traffic simulation. Our approach utilizes metaheuristic methods to create an analytical tool to assess vehicle performance. Secondly, we show how microscopic simulation run outputs can be post-processed to obtain vehicle and trip level performance metrics. The methodology will form the basis for a data-driven approach to unearthing trip experiences as realized by drivers in the real world. The methodology will contribute to, A.) Using vehicle trajectory traces to identify underlying vehicle maneuver distributions as obtained from real-world driver data, B.)

A robust Vehicle Model Architecture (VMA) has been developed to support model-based Vehicle System Control (VSC) design work and, in general, model-based vehicle system engineering activities. It is based on a logical breakdown of the vehicle into key subsystems with supporting bus infrastructure for distribution of signals between subsystems. Primary physical interfaces between the top level subsystems have been defined. Subsystem models that comply with these interfaces can be easily plugged into the architecture for complete simulation of vehicle systems. The VMA encourages model re-use and sharing between project teams and, furthermore, removes key obstacles to sharing of models with suppliers.

Vehicle dynamics CAE capabilities has increased in the past few years, specially, for handling and steering attributes. However, secondary ride simulations are still highly depended on the tire model. Such tire model must be capable to simulate high order phenomenon such as impact and harshness transmissibility in three directions. In order to gather tire information sufficient to cope with these phenomena, one needs to perform a series of specific tests, and so be able to build the intended tire model. Still, there could be correlation issues. This whole process takes a lot of time and resources. This article presents a semi-analytical approach, using data gathered via wheel force transducers (WFTs) that are typically used for load cascading and durability purposes. The method main advantage is that since it relies on measured data at the wheel center, it is independent of a tire model, and, as such, it demands less time and resources.

This paper describes the issues encountered in using conventionally acquired tire test data for dynamic total vehicle handling simulations and the need for improved methodology. It describes the new test procedure that was used to acquire all three forces and three moments in a transient mode for a matrix of loads, slip and camber angles. A study of the test data supports the premises that the overturning moment, Mx, should not be neglected in dynamic simulations, and that the effects of camber should not be treated as only an independent, linearly additive, camber thrust. Instead of the conventional application of a bi-cubic regression fit to a six region data division, a new algorithm is applied. The data is divided differently into five regions in the α - Fz plane, and a variable format regression equation is applied as appropriate. The resulting regression coefficients matrix is readily usable in dynamic simulations, and is shown to have a superior curve fit to the test data.

Yaw rate of a vehicle is highly influenced by the lateral forces generated at the tire contact patch to attain the desired lateral acceleration, and/or by external disturbances resulting from factors such as crosswinds, flat tire or, split-μ braking. The presence of the latter and the insufficiency of the former may lead to undesired yaw motion of a vehicle. This paper proposes a steer-by-wire system based on fuzzy logic as yaw-stability controller for a four-wheeled road vehicle with active front steering. The dynamics governing the yaw behavior of the vehicle has been modeled in MATLAB/Simulink. The fuzzy controller receives the yaw rate error of the vehicle and the steering signal given by the driver as inputs and generates an additional steering angle as output which provides the corrective yaw moment.

For realistic traffic modeling, real-world traffic calibration data is needed. These data include a representative road network, road users count by type, traffic lights information, infrastructure, etc. In most cases, this data is not readily available due to cost, time, and confidentiality constraints. Some open-source data are accessible and provide this information for specific geographical locations, however, it is often insufficient for realistic calibration. Moreover, the publicly available data may have errors, for example, the Open Street Maps (OSM) does not always correlate with physical roads. The scarcity, incompleteness, and inaccuracies of the data pose challenges to the realistic calibration of traffic models. Hence, in this study, we propose an approach based on spatial interpolation for addressing sparsity in vehicle count data that can augment existing data to make traffic model calibrations more accurate.

This paper proposes a low-cost but indirect method for occupancy detection and occupant counting purpose in current and future automotive systems. It can serve as either a way to determine the number of occupants riding inside a car or a way to complement the other devices in determining the occupancy. The proposed method is useful for various mobility applications including car rental, fleet management, taxi, car sharing, occupancy in autonomous vehicles, etc. It utilizes existing on-board motion sensor measurements, such as those used in the vehicle stability control function, together with door open and closed status. The vehicle’s motion signature in response to an occupant’s boarding and alighting is first extracted from the motion sensors that measure the responses of the vehicle body. Then the weights of the occupants are estimated by fitting the vehicle responses with a transient vehicle dynamics model.

With the emergence of Driving Automation Systems (SAE levels 1-5), the necessity arises for methods of evaluating these systems. However, these systems are much more challenging to evaluate than traditional safety features (SAE level 0). This is because an understanding of the Driving Automation system’s response in all possible scenarios is desired, but prohibitive to comprehensively test. Hence, this paper attempts to evaluate one such system, by modeling its behavior. The model generated parameters not only allow for objective comparison between vehicles, but also provide a more complete understanding of the system. The model can also be used to extrapolate results by simulating other scenarios without the need for conducting more tests. In this paper, low speed automated driving (also known as Traffic Jam Assist (TJA)) is studied. This study focused on the longitudinal behavior of automated vehicles while following a lead vehicle (LV) in traffic jam scenarios.