Search Results

Recognizing road conditions using onboard sensors is significant for the performance of intelligent vehicles, and the road profile is a widely accepted representation both in the temporal and frequency domains, greatly influencing driving quality. In this paper, a recurrent neural network embedded with attention mechanisms is proposed to reconstruct the road profile sequence. Firstly, the road and vehicle sensor signals are obtained in a simulated environment by modeling the road, tire, and vehicle dynamic system. After that, the models under different working conditions are trained and tested using the collected data, and the attention weights of the trained model are then visualized to optimize the input channels. Finally, field experiments on the real vehicle are conducted to collect real road profile data, combined with vehicle system simulation, to verify the performance of the proposed method.

The proposed Euro 7 regulation includes On Board Monitoring, or OBM, to continuously monitor vehicles for emission exceedances. OBM relies on feedback from existing or additional sensors to identify high emitting vehicles, which poses many challenges. Currently, sensors are not commercially available for all emissions constituents, and the accuracy of available sensors is not capable enough for in use compliance determination. On board emissions models do not offer enough fidelity to determine in use compliance and require new complex model innovation development which will be extremely complicated to implement on board the vehicle. The stack up of multi-component deterioration leading to an emissions exceedance is infeasible to detect using available sensors and models.

Complex chassis systems operate in various environments such as low-mu surfaces and highly dynamic maneuvers. The existing metrics for lateral motion hazard by Neukum [13] and Amberkar [17] have been developed and correlated to driver behavior against disturbances on straight line driving on a dry surface, but do not cover low-mu surfaces and dynamic driving scenarios which include both linear and nonlinear region of vehicle operation. As a result, an improved methodology for evaluating vehicle yaw dynamics is needed for safety analysis. Vehicle yaw dynamics safety analysis is a methodical evaluation of the overall vehicle controllability with respect to its yaw motion and change of handling characteristic.

The subsystem of front of dash (FOD) and instrument panel (IP) is a critical path to isolate the powertrain noise and road noise for vehicles. This subsystem mainly consists of sheet metal, dash mats, IP, and the components inside IP such as HVAC and wiring harness. To achieve certain level of cabin quietness, the sound transmission loss performance of this subsystem is usually used as a quantifier. In this paper, the sound transmission loss through the FOD and IP is investigated up to 10kHz, through both acoustic testing and numerical simulation. In the acoustic testing, the subsystem is cut from a vehicle and installed on the wall of two-rooms STL testing suite, with source room being reverberant and receiver room being anechoic. In the testing, various scenarios are measured to understand the contributions from different components.

General Motors (GM) is working towards a future world of zero crashes, zero emissions and zero congestion. It’s “Ultium” platform has revolutionized electric vehicle drive units to provide versatile yet thrilling driving experience to the customers. Three variants of traction power inverter modules (TPIMs) including a dual channel inverter configuration are designed in collaboration with LG Magna e-Powertrain (LGM). These TPIMs are integrated with other power electronics components inside Integrated power electronics (IPE) to eliminate redundant high voltage connections and increase power density. The developed power module from LGM has used state-of-the art sintering technology and double-sided cooled structure to achieve industry leading performance and reliability. All the components are engineered with high level of integration skills to utilize across TPIM variants.

With the advent of this new era of electric-driven automobiles, the simulation and virtual digital twin modeling world is now embarking on new sets of challenges. Getting key insights into electric motor behavior has a significant impact on the net output and range of electric vehicles. In this paper, a complete 3D CFD model of an Electric Motor is developed to understand its churning losses at different operating speeds. The simulation study details how the flow field develops inside this electric motor at different operating speeds and oil temperatures. The contributions of the crown and weld endrings, crown and weld end-windings, and airgap to the net churning loss are also analyzed. The oil distribution patterns on the end-windings show the effect of the centrifugal effect in scrapping oil from the inner structures at higher speeds. Also, the effect of the sump height with higher operating speeds are also analyzed.

Range anxiety in current battery electric vehicles is a challenging problem, especially for commercial vehicles with heavy payloads. Therefore, the development of electrified propulsion systems with multiple power sources, such as fuel cells, is an active area of research. Optimal speed planning and energy management, referred to as eco-driving, can substantially reduce the energy consumption of commercial vehicles, regardless of the powertrain architecture. Eco-driving controllers can leverage look-ahead route information such as road grade, speed limits, and signalized intersections to perform velocity profile smoothing, resulting in reduced energy consumption. This study presents a comprehensive analysis of the performance of an eco-driving controller for fuel cell electric trucks in a real-world scenario, considering a route from a distribution center to the associated supermarket.

Unmanned Aerial Vehicles (UAVs) encounter various uncertainties, including unfamiliar environments, signal delays, limited control precision, and other disturbances during task execution. Such factors can significantly compromise flight safety in complex scenarios. In this paper, to enhance the safety of UAVs amidst these uncertainties, a control accuracy prediction model based on ensemble learning abnormal state detection is designed. By analyzing the historical state data, the trained model can be used to judge the current state and obtain the command tracking control accuracy of the UAV at that instant. Ensemble learning offers superior classification capabilities compared to weak learners, particularly for anomaly detection in flight data. The learning efficacy of support vector machine, random forest classifier is compared and achieving a peak accuracy of 95% for the prediction results using random forest combined with adaboost model .

This paper proposes an intelligent car testing and evaluation method based on digital twins, which is crucial for ensuring the proper functioning of autonomous driving systems. This method utilizes digital twin testing technology to effectively map and integrate real vehicles in real-world testing scenarios with virtual test environments. By enriching the testing and validation environment for smart cars, this approach improves testing efficiency and reduces costs. This study connects real test vehicles with simulation software testing toolchains to build a digital twin autonomous driving testing platform. This platform facilitates the validation, testing, and evaluation of functional algorithms, and case study is conducted through testing and validation of an emergency collision avoidance system. By rapidly applying digital twin testing and evaluation techniques for intelligent cars, this approach accelerates the development and deployment of autonomous vehicles.

De-centralized brake actuation – that is, brake systems that incorporate individual actuators at each wheel brake location to both provide the apply energy and the modulation of braking force – is not a new area of study. Typically realized in the form of electro-mechanical brake calipers or drum brakes, or as “single corner” hydraulic actuators, de-centralized actuation in braking systems has already been deployed in production on General Motor EV1 Electric Vehicle (1997) in the form of electric drum brakes and has been studied continually by the automotive industry since then. It is frequently confused with “brake by wire,” and indeed practical implementations of de-centralized actuation are a form of brake by wire technology. However, with millions of vehicles on the road already with “brake by wire” systems - the vast majority of which have centralized brake actuation – the future of “brake by wire” is arguable settled.

Zinc-based electrogalvanized (EG) and hot-dip galvanized (HDGI) coatings have been widely used in automotive body-in-white components for corrosion protection. The formability of zinc coated sheet steels depends on the properties of the sheet and the interactions at the interface between the sheet and the tooling. The frictional behavior of zinc coated sheet steels is influenced by the interfacial conditions present during the forming operation. Friction behavior has also been found to deviate from test method to test method. In this study, various lubrication conditions were applied to both bending under tension (BUT) test and a draw bead simulator (DBS) test for friction evaluations. Two different zinc coated steels; electrogalvanized (EG) and hot-dip galvanized (HDGI) were included in the study. In addition to the coated steels, a non-coated cold roll steel was also included for comparison purpose.

Accelerating product development cycles and incentives to reduce costs in product development are strong motivators to move to virtual development and validation processes. Challenges to moving to a virtual paradigm include a wealth of historical data and context for hardware tests, uncertainty over dependencies, and a lack of a clear path of transition to virtual methods. In this paper we will discuss approaches to understanding the value created by hardware tests and aligning that value to virtual processes. We will also discuss the need for a virtual context to be added to SAE J1739 [1] (DFMEA detection criteria), and how to create paths to maximize the value of virtual assessments. Finally, we will also discuss the cultural and organizational changes required to support.

As passenger vehicle design evolves and accelerates, the use of multi-body dynamics solvers has proven to be invaluable in the engineering workflow. MBD solvers allow engineers to build virtual vehicle models that can accurately simulate vehicle responses and calculate internal forces, which previously could only be assessed using physical prototype builds with hundreds of measurement transducers. Evaluation and selection of solvers within an engineering environment is inherently a multi-dimensional activity that can include ease of use, retention of previously developed expertise, accuracy, speed, and integration with existing analysis processes. We discuss here some of the challenges present in developing capability and accumulating data to support each of these criteria. Developing a pilot model that is capable of being applied to a comprehensive set of use cases, and then verifying those use cases, required significant project management activity.

The development of intelligent and networked vehicles has enhanced the demand for precise road information perception. Detailed and accurate road surface information is essential to intelligent driving decisions and annotation of road surface semantics in high-precision maps. As one of the key parameters of road information, road roughness significantly impacts vehicle driving safety and comfort for passengers. To reach a rapid and accurate estimation of road roughness, in this study, we develop a neural network model based on vehicle response data by optimizing a long-short term memory (LSTM) network through the particle swarm algorithm (PSO), which fits non-linear systems and predicts the output of time series data such as road roughness precisely. We establish a feature dataset based on the vehicle response time domain data that can be easily obtained, such as the vehicle wheel center acceleration and pitch rate.

The interaction between driveline control and anti-lock braking system (ABS) control in electric vehicles (EV) was investigated based on multi-body dynamics (MBD) model and control model co-simulation. Two primary driveline control algorithms, active damping control and wheel flare control, were integrated with ABS control in Simulink model and the influence on ABS control was studied. The event for high mu to low mu transition was simulated. When ABS control is active on low mu surface, the vehicle shows large wheel slip and long duration time before wheel speed returns to stable control. This performance could be improved with activating driveline control. Deceleration uniformity metric shows that active damping control has very small effect when ABS control becomes stable after passing through the high mu to low mu transition period. Driveline damping control can help to reduce vibration, but it is difficult to find satisfied tuning for wheel speed performance.

Torque ripple from electric motor can excite a system resonance perceived as vibration at the vehicle seat track. The CAE simulation procedure was applied to analyze the seat track acceleration excited by electric motor torque ripple. In this study, the transfer function between the electric motor torque and vehicle level seat track acceleration was developed, and it incorporates the control capability and vehicle sensitivity subfunctions. The motor level torque ripple requirement was developed, which can support motor design in early vehicle development stage based on vehicle level criteria. The analysis results obtained for motor level torque ripple requirement shows good agreement with the experimental validation using vehicle test data. The variation study on control capability and vehicle sensitivity was investigated, and the results can help to identify the solution to improve vehicle torque ripple response.

Vehicle manufacturers conduct tests to develop crash sensing system calibrations. Ditch fall-over is one of a suite of laboratory tests used to develop rollover sensing calibrations that can trigger deployment of safety devices like roof rail airbags and seat belt pretensioners. The ditch fall-over test simulates a flat road followed by a ditch on one side of the road. The vehicle heads into the ditch and the driver applies swift steer input once the ditch slope is sensed. Typically, the steer input is applied when the two down-slope wheels on the ditch side enter the ditch. Multi-Body Dynamics (MBD) software can be used for virtual simulation of these test events. Conventionally in simulations, the vehicle-model is run without steer input and the marking line crossing time is observed/manually recorded from observation of simulation video. This recorded time is used to apply the steer input and the full event is then re-simulated.

The Continuous Fluid Level and Quality Indicator (CFLQI) technology is focused on increasing the sampling frequency of brake fluid reservoir volume and detecting specific brake fluid contaminants. CFLQI targets to improve diagnostics detection range and resulting degraded vehicle operation strategies by increasing sensitivity to brake fluid loss and the addition of a fluid quality feature. The theory of CFLQI is to improve future autonomous and highly automated vehicle performance, both of which will have reduced driver input and service schedules, by providing earlier fluid level and fluid health warnings. The two technologies selected to prove theory of operation were ultra-sonic sensor and capacitive sense element technology. Both technologies show initial capability to meet fluid sensing targets with system level ASIL D ASIC design. The CFLQI compliments and improves upon current technology of brake pad wear sensors, leak detection diagnostics and brake fluid level monitoring.

The validation of brake discs has remained, to this day, heavily reliant on “Thermal Abuse” or “Thermal Cracking” type testing, with many procedures so dated that most engineers active in the industry today cannot even recall the origin of the test. These procedures - of which there are many variants - all share the trait of greatly accelerating durability testing by performing repeated high power (high speed and high deceleration) brake applies to drive huge temperature gradients and internal stress, and often allowing the disc to get very hot, to where the strength of the material from which the disc is constructed is significantly degraded. There is little debate about whether these procedures work; by and large disc durability issues in the field are extremely rare.

Regenerative braking without question greatly impacts brake pad service life in the field, in most cases extending it significantly. Estimating its impact precisely has not been an overriding concern - yet - due in part to the extensive sharing of brake components between regen-intensive battery-electric and hybrid vehicles, and their more friction-brake intensive internal combustion engine powered sibling. However, a multitude of factors are elevating the need for a more accurate estimation, including the emerging of dedicated electric vehicle architectures with opportunities for optimizing the friction brake design, a sharp focus on brake particulate emissions and the role of regenerative braking, a need to make design decisions for features such as corrosion protection for brake pad and pad slide components, and the emergence of driver-facing features such as Brake Pad Life Monitoring.