Refine Your Search

Search Results

Technical Paper

EURO-NCAP MPDB Compatibility Impact Model Assessment Using a Virtual Barrier Deformation Tracker

2021-04-06
2021-01-0834
Euro NCAP committee has created the MPDB “Compatibility” test that could change the way we design the vehicle front structure for impact. To assist the crashworthy design development activity for this new mode of impact test, CAE MPDB barrier models have been developed and used by vehicle safety CAE engineers. However, to evaluate the “Standard Deviation” and “Bottom-Out” modifiers accurately from the CAE model run results, a virtual scanner, which can emulate the measurement accuracy of the physical scanner in measuring the depth and size of the deformed MPDB barrier, is required. Currently, the model is built with more than around 350,000 elements and hence the prohibitively large number of elements and nodes need to be tracked to reconstruct the barrier’s deformation contour which is necessary for the “Standard Deviation” and “Bottom-Out” assessment from vehicle-MPDB barrier impact model results.
Technical Paper

Optimum Engine Power Point Determination Method to Maximize Fuel Economy in Hybrid Vehicles

2021-04-06
2021-01-0419
One of the advantages of hybrid vehicles is the ability to operate the engine more optimally at a low brake specific fuel consumption (BSFC) as compared to conventional vehicles. This ability of hybrid vehicles is a major factor contributing to the fuel economy improvement over conventional vehicles. Unlike conventional gasoline powertrains, hybrid powertrains allow engine to be switched off and use battery power to propel vehicles. In order to maintain battery state of charge neutral operation between the start and end of a drive cycle, the net electrical energy consumption from the battery requires to be zero. An optimization algorithm can be developed and calibrated in different ways to achieve net zero battery energy over the cycle. For instance, the engine can be operated at powers higher than the power of the drive cycle to charge the battery. This accumulated energy can be used for all-electric propulsion by turning off the engine.
Technical Paper

Using A Representative Driving Pattern Extraction Technique Modeling with Machine Learning Development of Durability Test Mode

2021-04-06
2021-01-0160
Test mode for durability of powertrain often defines a method by reflecting figures such as frequency of use or severity, but in the complex systems, it is hard to verify durability in real environments with the simple conditions. Thus, this session presents the new analysis method that was modelled to apply machine learning techniques to extract representativel driving patterns from the perspective of powertrain loads reflecting the driving situation and driver will, and to develop realistic endurance test evaluation modes.
Technical Paper

Research on Vehicle Lane Change Based on Vehicle Speed Planning

2021-04-06
2021-01-0162
Lane changing manoeuvers is an essential rudiment in vehicle driving and has a significant impact on the characteristics of traffic flow. In the case of traditional cars, the driver operates the vehicle to complete the lane change whilst for autonomous vehicles, completing the lane change requires planning the lane change trajectory and controlling the vehicle speed during the lane change. Unreasonable lane change trajectory and vehicle speed may cause the vehicle to lose stability, threaten driving safety, increase energy consumption and waste energy. This paper considers the safety and economy of the lane changing process, and proposes a new lane changing method for vehicles.
Technical Paper

Dynamic Speed Limit for Self-Identifying Platoons of Mixed Vehicular Traffic on Freeways under Connected Environment

2021-04-06
2021-01-0168
Freeways and highway roads comprise about 200,000 of 4-million-mile public road network where a total of 5.3 trillion miles are traveled each year. Due to high contribution to mobility and energy consumption, freeways and highway have been attracting researchers to move more vehicles faster and in an energy-efficient manner. This includes developing better control methods able to sense real-time traffic, environmental factors, and road network characteristics. Traditional controls are vehicle actuated where the sensors are primarily fixed and have limited detecting capabilities. With the advent of connected vehicles (CVs), fixed sensor technology is being augmented with vehicle-based sensing. Moreover, pedestrians through their smart devices and non-motorized vehicles such as scooters, bikes are also being connected to get a better insight into the state of traffic.
Technical Paper

A Rapid Compression Machine Study on Ignition Delay Times of Gasoline Mixtures and Their Multicomponent Surrogate Fuels Under Diluted and Undiluted Conditions

2021-04-06
2021-01-0554
In this work autoignition delay times of two multi-component surrogates (high and low RON) were experimentally compared with their target full blend gasoline fuels. The study was conducted in a rapid compression machine (RCM) test facility and a direct test chamber (DTC) charge preparation approach was used for mixture preparation. Experiments were carried over the temperature range of 650K-900K and at 10 bar and 20 bar compressed pressure conditions for equivalence ratios of (Φ =) 0.6-1.3. Dilution in the reactant mixture was varied from 0% to 30% CO2 (by mass), with the O2:N2 mole ratio fixed at 1:3.76. This dilution strategy emulates exhaust gas recirculation (EGR) substitution in spark ignition (SI) engines. The multicomponent surrogate captured the reactivity trends of the gasoline-air mixtures reasonably well in comparison to the single component (iso-octane) surrogate.
Technical Paper

FCA US LLC - Magnesium Closures Development

2021-04-06
2021-01-0278
This paper will focus on automotive development highlights of Fiat Chrysler Automobiles US LLC (FCA US LLC) magnesium intensive closures components. FCA has a long tradition of innovation. Deploying lightweight materials is one of many key technologies that FCA US has implemented to reduce vehicle mass and improve overall fuel economy while maintaining rigorous functional objective performance. This paper will outline some basic design and manufacturing considerations for magnesium closures. The development of the 2017 Chrysler Pacifica Liftgate and 2018 Jeep Wrangler Swingate along with the 2nd generation magnesium spare tire bracket will be the focus.
Technical Paper

Assessing the Access to Jobs by Shared Autonomous Vehicles in Marysville, Ohio: Modeling, Simulating and Validating

2021-04-06
2021-01-0163
Autonomous vehicles are expected to change our lives with significant applications like on-demand, shared autonomous taxi operations. Considering that most vehicles in a fleet are parked and hence idle resources when they are not used, shared on-demand services can utilize them much more efficiently. While ride hailing of autonomous vehicles is still very costly due to the initial investment, a shared autonomous vehicle fleet can lower its long term cost such that it becomes economically feasible. This requires the Shared Autonomous Vehicles (SAV) in the fleet to be in operation as much as possible. Motivated by these applications, this paper presents a simulation environment to model and simulate shared autonomous vehicles in a geo-fenced urban setting.
Technical Paper

Assessing the Impacts of Dedicated CAV lanes in a Connected Environment: An application of Intelligent Transport Systems in Corktown, Michigan

2021-04-06
2021-01-0177
The interaction of Connect and Automated vehicles (CAV) with regular vehicles in the traffic stream has been extensively researched. Most studies, however, focus on the calibrating driver behavior models for CAVs based on various levels of automation and driver aggressiveness. Other related studies largely focus on the coordination of CAVs and infrastructure like traffic signals to optimize traffic. However, the effects of different strategic orchestrations of CAVs in the traffic stream in the comparative scenario-based analysis is understudied. Thus, this study develops a framework and simulations for integrating CAVs in a corridor section. We develop a calibrated model with CAVs for a corridor section in Corktown, Michigan, and simulate how dedicated CAV lane operations can be implemented without significant change in existing infrastructure.
Technical Paper

A Hybrid method for Automotive Entity Recognition

2021-04-06
2021-01-0179
Over the past decades, automotive industry has made substantial investments in automation solutions, electric and autonomous vehicles and advanced product technologies for enhancing vehicular communication and so on. The rise of industry 4.0 brings out a revolutionary transformation in the automotive industry with low-cost computing, high-speed connectivity, and machine learning that have enabled the digitization of the physical world, transforming insights into optimized actions. These technologies have an important role in the growth and future of automotive domain. Hence it is relevant and important to get insight of different OEMs (Original Equipment Manufacturer) at different geographical locations and their focusing technologies adapted currently and in the nearby future. In this paper, we are implementing a hybrid method for entity recognition, which is a combination of both rule-based and machine learning based entity recognition techniques.
Technical Paper

A comparative study of physics based grey-box and neural network trained black-box dynamic models in an RCCI engine control parameter prediction

2021-04-06
2021-01-0178
Real time in-cylinder based parameters such as start of combustion (Ɵsoc), 50% fuel mass burn crank angle (Ɵ50), burn duration (Ɵbd), indicated mean effective pressure (IMEP) etc. are considered as very challenging engine control related information in reactivity controlled compression ignition (RCCI) engines. In advanced engine control architecture, direct cylinder pressure sensor (CPS) signals are internally computed in observer module to estimate these parameters. However, CPS is expensive and demands a bulky engine hardware. Lately, physics based control models or grey-box models in RCCI engines were considered as a cost competitive and smart alternative to hardware based signal source. But, physics based models are complex and are restricted to a very few computation worthy parameters only. In this context, artificial neural networks (ANN) presents a viable alternative to increase the number of control worthy parameters and complement physics based models.
Technical Paper

Predictive Gearbox Oil temperature using Machine Learning

2021-04-06
2021-01-0182
Gearbox is one of the most defining components for vehicles, turbines and other applications. A failure in the gearbox would ultimately cause the system to breakdown and thus results in operational failure. A gearbox failure can be attributed to several factors such as gearbox oil temperature, driving patterns, dependent engine components and other various gearbox performance.The focus of this paper is gearbox oil temperature sensor which is the one important factor that determine gearbox overheating and influence the system to take precautionary steps in switching from different types of oil to prevent risk of damaging their equipment and expensive repair .The goal of this paper is to predict the gearbox oil temperature sensor failure by adopting machine learning techniques.Various machine learning techniques such as Support vector machine, decision trees and random forest etc are employed in this paper to achieve the objective.
Technical Paper

High-Fidelity Modeling of Light-Duty Vehicle Emissions and Fuel Economy Using Deep Neural Networks

2021-04-06
2021-01-0181
The transportation sector contributes around 28% of all emissions and air pollution globally. Emission models of modern vehicles are great tools to estimate the impact of technology on vehicle emission reduction but developing a simple and high fidelity model is challenging due to the variety of vehicle classes, driving conditions, driver behaviors, and other physical and operational constraints. But, recent literature indicates that neural network based models may address these concerns due to high computation speed and high-accuracy of predicted emissions. In this study we seek to expand upon this initial research by utilizing several deep neural networks (DNN) models such as LSTM and CNN. These DNN algorithms are developed and a comprehensive assessment of their performances is done. Sensitivity analysis is carried out for input parameters selection. Training and testing datasets are selected and cross-validation is done for validation of the learning procedure.
Technical Paper

Machine Learning and Response Surface-Based Numerical Optimization of the Combustion System for a Heavy-duty Gasoline Compression Ignition Engine

2021-04-06
2021-01-0190
In this study, the combustion system of a heavy-duty diesel engine was optimized towards gasoline compression ignition (GCI) mode using computational fluid dynamics (CFD), response surface methodology (RSM), and machine learning (ML). The targets for optimization include the combustion bowl geometry, the injector specifications, and the swirl ratio. The optimization starts with a CFD design of experiment campaign, where 128 concepts were generated by a parametric combustion recipe design process and were evaluated using CONVERGE CFD solver. Performance was evaluated across three load points to identify an optimal combustion system design. Subsequently, two different approaches were used to further optimize the combustion system design. The first approach is the traditional RSM-based optimization and the second approach is a machine learning-based optimization strategy.
Technical Paper

Scenario Uncertainty Modeling for Predictive Maintenance with Recurrent Neural Adaptive Processes (RNAP)

2021-04-06
2021-01-0191
For commercial-vehicle original equipment manufacturers, predictive maintenance has drawn attention for the benefits to facilitate money-saving and increased road safety. Data-driven models have been widely explored and implemented as predictive maintenance solutions. However, the working scenarios for different commercial-vehicles vary a lot, which makes it difficult to build a universal model suitable for all the cases. In this paper, we propose a Recurrent Neural Adaptive Processes network to adapt to different scenarios by modeling the uncertain at the same time. The ensemble network combines the traits of neural processes, recurrent neural network and meta learning together. Neural processes could achieve prediction in stochastic based on the input dataset. Meta-learning is good at dealing with few-shot multi-tasks learning, and recurrent networks are utilized as the encoder of the neural processes to fit for the input time-series data.
Technical Paper

Analyzing the impact of Electric Vehicles Charging Stations on Power Quality in Power Distribution System

2021-04-06
2021-01-0199
Electric vehicles (EVs) have become one of the promising solutions to decrease the impact of fossil fuel combustion engines on the environment. In spite of their significant role in decreasing air pollutants, EVs could have adversarial impacts on quality of power in electric grids. Electric vehicles use rechargeable batteries to store the energy and run their electric engines. EVs’ battery chargers use power electronic switches to convert AC voltage to DC voltage, leading to power quality issues due to its non-linearity. Non-linear loads have negative impacts on power quality and can cause several problems such as voltage unbalance, voltage fluctuations, low power factors, and harmonics in power distribution systems. Charging a high number of electric vehicles will create power quality issues and cause harmonic distortions. Harmonics affect the performance of power transformers by increasing power losses and reducing their output power.
Technical Paper

Research on Automatic Joint Calibration Method of Multi 3D-LIDARs and Inertial Measurement Unit

2021-04-06
2021-01-0070
In the field of automatic driving, the combination of 3D LIDAR and inertial measurement unit (IMU) is a common sensor configuration scheme in laser point-cloud localization, high-precision map making and point-cloud target detection. So it is critical to calibrate LIDAR and IMU accurately. At present, due to the large volume and high cost of 3D LIDAR with high-line-number(Such as 64 lines or 128 lines), the configuration scheme of using multiple low-line-number 3D LIDARs appears in the automatic driving vehicle sensing system. However, the common calibration methods are not suitable for multi 3D LIDARs and IMU parameters calibration, which have the disadvantages of cumbersome implementation and low accuracy. In this paper, a joint calibration test platform composed of dual LIDARs and IMU is assembled, and we propose the corresponding automatic calibration method.
Technical Paper

A Real-time Curb Detection Method for Vehicle by Using a 3D-LIDAR Sensor

2021-04-06
2021-01-0076
Effectively detecting road boundaries in real time is critical to the applications of autonomous vehicles, such as vehicle localization, path planning and environmental understanding.To precisely extract the road boundaries from the 3D LiDAR data, a real-time curb detection algorithm consisting of four steps is proposed in this paper.
Technical Paper

Accurate Pressure Control based on Driver Braking Intention Identification for a Novel Integrated Braking System

2021-04-06
2021-01-0100
With the development of intelligent and electric vehicles, higher requirements are put forward for the active braking and regenerative braking ability of the braking system. The traditional braking system equipped with vacuum booster has difficulty meeting the demand, therefore it has gradually been replaced by the Integrated Braking System. In this paper, a novel integrated braking system is presented, which mainly contains a pedal feel simulator, a permanent magnet synchronous motor(PMSM), a series of transmission mechanisms, and the hydraulic control unit. As an integrative system of mechanics-electronics-hydraulics, the Integrated Braking System has complex nonlinear characteristics, which challenge accurate pressure control. Furthermore, it is a completely decoupled braking system, the pedal force doesn’t participate in pressure-building, so it is necessary to precisely identify driver’s braking intention.
Technical Paper

Object Detection and Augmented Visualization based on Panoramic Image Segmentation

2021-04-06
2021-01-0089
Panoramic images can provide critical information for advanced driving assistance systems (ADAS), such as parking slots and surrounding vehicles. However, the vehicle in the bird's-eye view image is severely deformed and incomplete, the parking slots are easily shielded, and the visual information becomes very blurred in some insufficiently illuminated environments. When the driver cannot see the surrounding environment information clearly, the risk of collision will increase, especially during parking. To solve the problem of local environment perception based on panoramic images, we use the results of panoramic image segmentation to construct a virtual surround view monitoring system for the driver to observe the surrounding environment directly and clearly. Firstly, a lightweight segmentation network is redesigned based on SegNet, and it will improve the segmentation accuracy without increasing the inference time.
X