Search Results

This paper compares the results from three human factors studies conducted in a motion-based simulator in 2008, 2014 and 2023, to highlight the trends in driver's response to Forward Collision Warning (FCW). The studies were motivated by the goal to develop an effective HMI (Human-Machine Interface) strategy that enables the required driver's response to FCW while minimizing the level of annoyance of the feature. All three studies evaluated driver response to a baseline-FCW and no-FCW conditions. Additionally, the 2023 study included two modified FCW chime variants: a softer FCW chime and a fading FCW chime. Sixteen (16) participants, balanced for gender and age, were tested for each group in all iterations of the studies. The participants drove in a high-fidelity simulator with a visual distraction task (number reading). After driving 15 minutes in a nighttime rural highway environment, a surprise forward collision threat arose during the distraction task.

Identifying edge cases for designed algorithms is critical for functional safety in autonomous driving deployment. In order to find the feasible boundary of designed algorithms, simulations are heavily used. However, simulations for autonomous driving validation are expensive due to the requirement of visual rendering, physical simulation, and AI agents. In this case, common sampling techniques, such as Monte Carlo Sampling, become computationally expensive due to their sample inefficiency. To improve sample efficiency and minimize the number of simulations, we propose a tailored active learning approach combining the Support Vector Machine (SVM) and the Gaussian Process Regressor (GPR). The SVM learns the feasible boundary iteratively with a new sampling point via active learning. Active Learning is achieved by using the information of the decision boundary of the current SVM and the uncertainty metric calculated by the GPR.

The automotive industry is passing for a big transformation, due to technologies advance. The electrical technologies are also on a good rising curve, calling the attention of the Original Equipment Manufacturer (OEMs). This scenario generates the demand for a faster method to train their new hired engineers, when compared with usual on the job training. Model in the Loop (MiL) consists in one of the real-time embedded systems test phases, which is developed in a computational environment, performing a mathematical modeling of the system, presenting an interface that allows the visualization of its dynamics and the signals involved. Two powerful software in industry that apply MiL are the Matlab and Simulink. A project involving these applications was proposed for a team of new hired engineers, developing models of several vehicle Electronic Control Units (ECUs), with some scope reduction as an example the functional requirements reduction.

This paper focuses on finding and analyzing the relevant parameters affecting traffic flow when autonomous vehicles are introduced for ride hailing applications and autonomous shuttles are introduced for circulator applications in geo-fenced urban areas. For this purpose, different scenarios have been created in traffic simulation software that model the different levels of autonomy, traffic density, routes, and other traffic elements. Similarly, software that specializes in vehicle dynamics, physical limitations, and vehicle control has been used to closely simulate realistic autonomous vehicle behavior under such scenarios. Different simulation tools for realistic autonomous vehicle simulation and traffic simulation have been merged together in this paper, creating a realistic simulator with Hardware-in-the-Loop (HiL), Traffic-in-the-Loop (TiL), and Software in-the-Loop (SiL) simulation capabilities.

Real-world second-by-second vehicle driving cycle data is very important for vehicle research and development. A project solely dedicated to generating such information would be tremendously costly and time consuming. Alternatively, we developed such a database by utilizing two publicly available passenger vehicle travel surveys: 2004-2006 Puget Sound Regional Council (PSRC) Travel Survey and 2011 Atlanta Regional Commission (ARC) Travel Survey. The surveys complement each other - the former is in low time resolution but covers driver operation for over one year whereas the latter is in high time resolution but represents only one-week-long driving operation. After analyzing the PSRC survey, we chose 382 vehicles, each of which continuously operated for one year, and matched their trips to all the ARC trips. The matching is carried out based on trip distance first, then on average speed, and finally on duration.

The fuel consumption of a vehicle is shown to be linearly proportional to (1) total vehicle work required to drive the cycle due to mass and acceleration, tire friction, and aerodynamic drag and (2) the powertrain (PT) mechanical losses, which are approximately proportional to the engine displaced volume per unit distance travelled (displacement time gearing). The fuel usage increases linearly with work and displacement over a wide range of applications, and the rate of increase is inversely proportional to the marginal efficiency of the engine. The theoretical basis for these predictions is reviewed. Examples from current applications are discussed, where a single PT is used across several vehicles. A full vehicle cycle simulation model also predicts a linear relationship between fuel consumption, vehicle work, and displacement time gearing and agrees well with the application data.

Electric vehicles are highly sensitive to variations in environmental factors (like temperature, drive style, grade, etc.). The distribution of real-world range of electric vehicles due to these environmental factors is an important consideration in target setting. This distribution can be obtained by running several simulations of an electric vehicle for a number of high-frequency velocity, grade, and temperature real-world trip profiles. However, in order to speed up simulation time, a unique set of drive profiles that represent the entire real-world data set needs to be developed. In this study, we consider 40,000 unique velocity and grade profiles from various real-world applications in EU. We generate metadata that describes these profiles using trip descriptor variables. Due to the large number of descriptor variables when considering second order effects, we normalize each descriptor and use principal component analysis to reduce the dimensions of our dataset to six components.

At various milestones during a vehicle’s development program, different CAE models are created to assess NVH error states of concern. Moreover, these CAE models may be developed in different commercial CAE software packages, each one with its own unique advantages and strengths. Fortunately, due to the wide spread acceptance that the Functional Mock-up Interface (FMI) standard gained in the CAE community over the past few years, many commercial CAE software now support cosimulation in one form or the other. Cosimulation allows performing multi-domain/multi-resolution simulations of the vehicle, thereby combining the advantages of various modeling techniques and software. In this paper, we explore cosimulation of full 3D vehicle model developed in MSC ADAMS with 1D driveline model developed in LMS AMESim. The target application of this work is investigation of vehicle NVH error states associated with both hybridized and non-hybridized powertrains.

Modeling of High Voltage (HV) wires is an important aspect of vehicle safety simulations for electrified powertrains to understand the potential tearing of the wire sheath or pinching of HV wiring. The behavior of the HV wires must be reviewed in safety simulations to identify potential hazards associated with HV wire being exposed, severed, or in contact with ground planes during a crash event. Modeling HV wire is challenging due to the complexity of the physical composition of the wire, which is usually comprised of multiple strands bundled and often twisted together to form the HV electrical conductor. This is further complicated by the existence of external insulating sheathing materials to prevent HV exposure during normal operating conditions. This paper describes a proposed method to model and characterize different types of HV wires for usage in component- and vehicle-level safety models.

The potential for Augmented Reality (AR) spans many domains. Among other applications, AR can improve the discovery and learning experience for users inspecting a particular item. This paper discusses the use of AR in the automotive context; particularly, on improving the user experience in a dealership show room. Visual augmentation, through a tablet computer or glasses allows users to take part in a self-guided tour in learning about the various features, details, and options associated with a vehicle. The same approach can be applied to other learning scenarios, such as training and maintenance assistance. We evaluated a set of AR Glasses and a general purpose tablet. A table-top showroom was developed demonstrating what the actual user experience would be like for a self-guided dealership tour using natural markers and three-dimensional content spatially registered to physical objects in the user’s field of view.

Software for autonomous vehicles is highly complex and requires vast amount of vehicle testing to achieve a certain level of confidence in safety, quality and reliability. According to the RAND Corporation, a 100 vehicle fleet running 24 hours a day 365 days a year at a speed of 40 km/hr, would require 17 billion driven kilometers of testing and take 518 years to fully validate the software with 95% confidence such that its failure rate would be 20% better than the current human driver fatality rate [1]. In order to reduce cost and time to accelerate autonomous software development, Hardware-in-the-Loop (HIL) simulation is used to supplement vehicle testing. For autonomous vehicles, path following controls are an integral part for achieving lateral control. Combining the aforementioned concepts, this paper focuses on a real-time implementation of a path-following lateral controller, developed by Freund and Mayr [2].

Use of sensors to monitor dynamic performance of machine tools at Ford’s powertrain machining plants has proven to be effective. The traditional approach to convert sensor data to actionable intelligence consists of identifying single features from cycle based signatures and setting thresholds above acceptable performance limits based on trials. The thresholds are used to discriminate between acceptable and unacceptable performance during each cycle and raise alarms if necessary. This approach requires a significant amount of resource & time intensive set up work up-front and considerable trial and error adjustments. The current state does not leverage patterns that might be discernible using multiple features simultaneously. This paper describes enhanced methods for processing the data using supervised and unsupervised machine learning methods. The objective of using these methods is to improve the prediction accuracy and reduce up-front set up.

Increasing number of vehicles are equipped with telematics devices and are able to transmit vehicle CAN bus information remotely. This paper examines the possibility of identifying individual drivers from their driving signatures embedded in these telematics data. The vehicle telematics data used in this study were collected from a small fleet of 30 Ford Fiesta vehicles driven by 30 volunteer drivers over 15 days of real-world driving in London, UK. The collected CAN signals included vehicle speed, accelerator pedal position, brake pedal pressure, steering wheel angle, gear position, and engine RPM. These signals were collected at approximately 5Hz frequency and transmitted to the cloud for offline driver identification modeling. A list of driving metrics was developed to quantify driver behaviors, such as mean brake pedal pressure and longitudinal jerk. Random Forest (RF) was used to predict driver IDs based on the developed driving metrics.

An engine cooling system in an automotive vehicle comprises of heat exchangers such as a radiator, charge air cooler and oil coolers along with engine cooling fan. Typical automotive engine-cooling fan assembly includes an electric motor mounted on a shroud that encloses the radiator core. One of main drivers of fan shroud design is Noise, Vibration, and Harshness (NVH) requirements without compromising the main function of airflow for cooling requirements. In addition, there is also a minimum stiffness requirement of fan shroud which is often overlooked in arriving at optimal design of it. Low Speed Damageability (LSD) assessment of an automotive vehicle is about minimizing the cost of repair of vehicle damages in low speed crashes. In low speed accidents, these fan motors are subjected to sudden decelerations which cause fan motors to swing forward thereby damaging the radiator core. So designing fan shroud for low speed damageability is of importance today.

The Pelvis-Thorax Side Air Bag (PTSAB) is a typical restraint countermeasure offered for protection of occupants in the vehicle during side impact tests. Currently, the dynamic performance of PTSAB for occupant injury assessment in side impact is limited to full-vehicle evaluation and sled testing, with limited capability in computer aided engineering (CAE). The widely used CAE method for PTSAB is a flat bag with uniform pressure. The flat PTSAB model with uniform pressure has limitations because of its inability to capture airbag deployment during gap closure which results in reduced accuracy while predicting occupant responses. Hence there is a need to develop CAE capability to enhance the accuracy of prediction of occupant responses to meet performance targets in regulatory and public domain side impact tests. This paper describes a new CAE methodology for assessment of PTSAB in side impact.

Pedestrian protection assessment methods require multiple head impact tests on a vehicle’s hood and other front end parts. Hood surfaces are often lifted up by using pyrotechnic devices to create more deformation space prior to pedestrian head impact. Assessment methods for vehicles equipped with pyrotechnic devices must also validate that the hood deployment occurs prior to head impact event. Estimation of pedestrian head impact time, thus, becomes a critical requirement for performance validation of deployable hood systems. In absence of standardized physical pedestrian models, Euro NCAP recommends a list of virtual pedestrian models that could be used by vehicle manufacturers, with vehicle FEA (Finite Element Analysis) models, to predict the potential head impact time along the vehicle front end profile. FEA simulated contact time is used as target for performance validation of sensor and pyrotechnic deployable systems.

Industry trends towards lighter, more aerodynamically efficient road vehicles have the potential to degrade a vehicle’s response to crosswinds. In this paper, a methodology is outlined that indirectly couples a computational fluid dynamics (CFD) simulation of the vehicle’s aerodynamic characteristics with a multi-body dynamics simulation (MBD) to determine yaw, roll and pitch response characteristics during a severe crosswind event. This one-way coupling approach mimics physical test conditions outlined in open loop test procedure ISO 12021:2010 that forms part of the vehicle sign-off criterion at Ford Motor Company. The methodology uses an overset mesh CFD method to drive the vehicle through a prescribed crosswind event, providing unfiltered predictions of vehicle force and moment responses that are used as applied forces in the MBD model. The method does not account for changes in vehicle attitude due to applied aerodynamic forces and moments.

Powertrain drivability evaluation and calibration is an important part of vehicle development to enhance the customer experience. This step mainly takes place on vehicle testing very late in the product development cycle, and is associated with a considerable amount of prototype, test facility, human resource and time cost. Design change options at this stage are also very limited. To reduce the development cost, a model based computer aided engineering (CAE) method is introduced and combined with hardware-in-the-loop (HIL) simulation technology. The HIL simulation method offers a possibility for drivability prediction and development in early phase of product cycle. This article describes the drivability HIL simulation process under development in Ford. The process consists of real time capable multi-domain CAE model integration, powertrain control module (PCM) and HIL simulator interface development and drivability HIL simulation.

Simulink is a very successful and popular method for modelling and auto-coding embedded automotive features, functions and algorithms. Due to its history of success, university feeder programs, and large third party tool support, it has, in some cases, been applied to areas of the software system where other methods, principles and strategies may provide better options for the software and systems engineers and architects. This paper provides approaches to determine when best to apply UML and when best to apply Simulink to a typical automotive feature. Object oriented software design patterns as well as general guidelines are provided to help in this effort. This paper's intent is not to suggest a replacement for Simulink but to provide the software architects and designers additional options when decomposing high level requirements into reusable software components.

Modern vehicles become increasingly software intensive. Software development therefore is critical to the success of the manufacturer to develop state of the art technology. Standards like ISO 26262 recommend requirement-based verification and test cases that are derived from requirements analysis. Agile development uses continuous integration tests which rely on test automation and evaluation. All these drove the development of a new model-based software verification environment. Various aspects had to be taken into account: the test case specification needs to be easily comprehensible and flexible in order to allow testing of different functional variants. The test environment should support different use cases like open-loop or closed-loop testing and has to provide corresponding evaluation methods for continuously changing as well as for discrete signals.