Browse Topic: Machine learning

Items (1,059)

Recommendation of Adaptive Safe Speed Limit for different driver Profiles using Deep Learning

2025-01-8106To be published on 04/01/2025

Personalization and driver profiling have emerged as popular topics within the Advanced Driver Assistance Systems domain. Several features, like seat position adjustment and route selection deliver a customized experience for everyone. The systems often profile the driver and their preferences using Machine Learning algorithms to anticipate the needs of the driver. Within this field, personalization of speed limits is not explored to the fullest. Currently, speed limits are the same for all drivers, irrespective of their driving style. Even when the speed limits are designed to address adverse weather conditions, they do not consider the driver’s capabilities. Thus, we propose a novel idea to recommend a safe speed limit based on the driving style and the surrounding environmental conditions. The architecture consists of Long Short-Term Memory based neural networks that capture the driver’s traits based on risks of the environment. The system identifies if a driver is uncomfortable in

Perumal, Rathapriya, Chouhan, Madhvendra, Rangarajan, Rishi

Deep Learning-Based Intelligent Tire Wear Detection Method

2025-01-8278To be published on 04/01/2025

In order to obtain real-time tire wear status of vehicles, this paper proposes a deep learning-based tire wear detection method. First, a PVDF piezoelectric film sensor is attached to the center of the tire airtight layer with different wear degrees to collect tire stress data under different working conditions. Secondly, the collected data is filtered, and the time domain feature information is extracted by principal component analysis to construct the feature data set. Finally, a deep regression model is established to train the feature data set to realize the real-time detection of tire damage. The results show that the tire wear detection error based on deep learning is less than 0.3mm, which has a high tire wear detection accuracy. It provides tire information for vehicle safe running and has high industrial application value

Xianyi, Xie, Yang, hao, jin, Lisheng

Lithium battery health status and lifespan prediction based on deep learning

2025-01-8126To be published on 04/01/2025

Rechargeable lithium batteries are widely used in the electric vehicle industry due to their long lifespan and high energy density. However, after long-term repeated charging and discharging, various electrochemical reactions inside lithium batteries can lead to performance degradation and even cause battery fires. Estimating the health status and predicting the remaining life of lithium batteries can provide insights into their future operating conditions, which is crucial for achieving fault warnings and ensuring the safe operation of battery-related equipment. In terms of predicting the health status of lithium batteries, this paper proposes a method based on an improved LSTM for health status estimation. This method first employs nearest neighbor component analysis to eliminate feature redundancy among the multidimensional health factors of the battery. Then, the differential grey wolf optimization algorithm is used to globally optimize the hyperparameters of the LSTM. The

K, meng zi

Application of AI/ML based Image Analytics in Auto Component Fracture Analysis

2025-01-8228To be published on 04/01/2025

This paper introduces an innovative digital solution for the categorization and analysis of fractures in Auto components, leveraging Artificial Intelligence and Machine Learning (AI/ML) technologies. The proposed system automates the fracture analysis process, enhancing speed, reliability, and accessibility for users with varying levels of expertise. The platform enables users to upload images of fractured parts, which are then processed by an AI/ML engine. The engine employs an image classification model to identify the type of fracture and a segmentation model to detect and analyze the direction of the fracture. The segmentation model accurately predicts cracks in the images, providing detailed insights into the direction and progression of the fractures. Additionally, the solution offers an intuitive interface for stakeholders to review past analyses and upload new images for examination. The AI/ML engine further examines the origin of the fracture, its progression pattern, and the

Sahoo, Priyabrata, Rawat, Sudhanshu, Garg, Vipin, Naidu, Garima, Sharma, Amit, Narula, Rahul, Bindra, Ritesh, Khera, Pankaj, Goel, Pooja, MONDAL, ARUP

A Fault Prediction Model for Electric Vehicle Charging Equipment Based on Adaptive Dynamic Thresholds

2025-01-8117To be published on 04/01/2025

The surge in electric vehicle usage has expanded the number of charging stations, intensifying demands on their operation and maintenance. Public charging stations, often exposed to harsh weather and unpredictable human factors, frequently encounter malfunctions requiring prompt attention. Current methods primarily employ data-driven approaches or rely on empirical expertise to establish warning thresholds for fault prediction. While these approaches are generally effective, the artificially fixed thresholds they employ for fault prediction limit adaptability and fall short in sensitivity to special scenarios, timings, locations, and types of faults, as well as in overall intelligence. This paper presents a novel fault prediction model for charging equipment that utilizes adaptive dynamic thresholds to enhance diagnostic accuracy and reliability. By integrating and quantifying Environmental Influence Factors (EF), Scenario Influence Factors (SF), Fault Severity Factors (FF), and

Wang, Hao, Wang, Ning, Li, Yuan, Tang, Xinyue

AI-Enhanced CAE Simulations: A Revolutionary Approach to Automotive Design and Engineering

2025-01-8241To be published on 04/01/2025

In the automotive industry, the durability and thermal analysis of components significantly impact vehicle component robustness and customer satisfaction. Traditional computer-aided engineering (CAE) methods, while effective, often involve extensive design iterations and troubleshooting, leading to prolonged development times and increased costs. The integration of artificial intelligence (AI) and machine learning (ML) into the CAE process presents a transformative solution to these challenges. By leveraging AI and ML, the durability simulation time of automobile components is significantly enhanced. Altair’s Physics AI tool utilizes historical CAE data to train ML models, enabling accurate predictions of model performance in terms of durability and stiffness. This reduces the necessity for multiple simulations, thereby decreasing CAE model design and solution completion times by 30%. By predicting potential issues early in the design phase, AI and ML allow engineers to make informed

Patil, Amol, Sonavane, Pravinkumar

Machine Learning Based Lane Detection and Lateral Offset Estimation Model for Vehicle Following Applications

2025-01-8020To be published on 04/01/2025

Precisely understanding the driving environment and determining the vehicle's exact position are crucial for the safe operation of Autonomous Vehicles. In the vehicle following systems that offer higher energy efficiency by precisely following a lead vehicle, the relative position of the ego vehicle to lane center is a key measure to a safe automated speed and steering control. This article presents a novel Enhanced Lane Detection technique with centimeter-level accuracy in estimating the vehicle offset from the lane center using the front-facing camera. Leveraging state-of-the-art computer vision models, the Enhanced Lane Detection technique utilizes YOLO-V8 image segmentation, trained on a diverse real-world driving scenarios dataset, to detect the driving lane. To measure the vehicle lateral offset, our model introduces a novel calibration method using nine reference markers aligned with the vehicle perspective and converts the lane offset from image coordinates to real-world

Karuppiah Loganathan, Nirmal Raja, Poovalappil, Aman, Naber, Jeffrey, Robinette, Darrell, Bahramgiri, Mojtaba

Tire wear life prediction using machine learning technique

2025-01-8757To be published on 04/01/2025

As global warming and environmental problems are becoming more serious, tires are required to accomplish balanced performance such as low rolling resistance, wet braking performance, driving stability, and ride comfort, while minimizing wear, noise, and weight. Furthermore, to respond to the rapid changes in the market environment, there is a strong demand for tire performance prediction and testing methods adapted to MBD. In MBD, Magic Formula and Ftire are used in performance design to predict handling and stability and ride comfort. However, predicting tire wear life, which is affected by both vehicle and tire characteristics, is difficult. And practical prediction method for estimating tire wear life has long been awaited. So, we had proposed an experimental-based tire wear life prediction method that using actual measurement tire characteristics and an abrasive wear volume formula. This method achieves accuracy that can be practically used in the early stages of vehicle

ANDO, TAKASHI

Highly accurate Machine Learning models for Automotive Crash applications using CAE centric AI/ML Platform

2025-01-8719To be published on 04/01/2025

In the automotive industry, there have been many efforts of late in using Machine Learning tools to aid virtual simulation and decrease product development time and cost. As the simulation world grapples with how best to incorporate ML techniques, two main challenges are evident. There is the risk of not knowing if the trained model is sound enough to drive the design. In addition, the complexity of porting simulation data back and forth to a ML software can make the process cumbersome. We would like to put forth a highly accurate and comprehensive one-stop ML workflow that has been implemented within a commercially available CAE driven platform called DEP MeshWorks. MeshWorks is used for rapid concept CAE and CAD model generation and parametrization. The AI platform within Meshworks leverages the inherent best-in-class parameterization capability to help rapidly generate data for training purposes. It is relatively quick to set up a fully parameterized CAE model and vary design

Krishnan, Radha

Load-Deflection Evaluation of Bump Stoppers Using a Machine Learning Approach

2025-01-8208To be published on 04/01/2025

The accurate prediction and evaluation of load-deflection behavior in bump stoppers are critical for optimizing driving performance and durability in the automotive industry. Traditional methods, such as extensive experimental testing and finite element analysis (FEA), are often time-consuming and costly. This paper introduces a machine learning-based approach to efficiently evaluate load-deflection curves for bump stoppers, thereby streamlining the design and testing process. By leveraging a comprehensive dataset that includes historical test results, material properties, and geometrical dimensions, various machine learning models, including Neural Networks, were trained to predict load-deflection behavior with high accuracy. This approach reduces the reliance on extensive physical testing and simulations, significantly enhancing the design optimization process, leading to faster development cycles and more precise performance predictions. A case study is presented to demonstrate the

Hazra, Sandip, Tangadpalliwar, Sonali

A Proactive System to Anticipate and Store the Damage Information on a Parked Car using Machine Learning Algorithm

2025-01-8211To be published on 04/01/2025

Vehicle ADAS Systems majorly comprises of two functions: Driving and Parking. The most common form of damage to the vehicle which goes unnoticed with unidentified cause are parking damages. A vehicle once parked at a certain location may get damaged without knowledge of the user. In this work developed a solution that not only pre-warns the driver but also prepares the vehicle beforehand if it suspects a damage may occur. This eliminates the latency between damage and information capture, detects small damages such as scratches, classifies the type of damage and also informs the user beforehand. This is solution is different from our competitors as the existing solutions informs the user about the scratches/damages, but these solutions are expensive, have high response time, and the damage information is captured after the damage has occurred. The solution consists of the following check blocks: Precondition, Sensor Control and Action Module. The Precondition Module observes the

Debnath, Sarnab, PATIL, Prasad, Belur Subramanya, Sheshagiri, Govinda, Shiva Prasad

Enhanced Secondary Crash Classification through Functional Class-Based Weighting and Hybrid Machine Learning Models

2025-01-8207To be published on 04/01/2025

Road safety and traffic management face significant challenges due to secondary crashes, which frequently cause more traffic, delays, and collisions. The importance of different types of roads is often overlooked by traditional methods for anticipating secondary crashes, which can result in suboptimal predictions and reaction plans. This research provides a novel method that combines a hybrid machine-learning model with a functional class-based weighting strategy to classify secondary crashes. The functional classes of the dataset are classified as interstates, arterial roads, collector roads, and local roads. The dataset also includes comprehensive crash narratives and various road attributes. Each functional class is assigned a weight that reflects its proportional importance in the likelihood of a subsequent crash, based on historical data and road usage patterns. After that, this weighting technique is integrated into a hybrid model architecture that trains a Random Forest model on

Patil, Mayur, Marik PE, Stephanie

Innovative Methods for Predicting Material Stress-Strain Curves Based on Transfer Learning and Artificial Intelligence

2025-01-8317To be published on 04/01/2025

The mechanical properties of materials play a crucial role in predicting performance across various applications. However, traditional methods for measuring these properties often involve complex, resource-intensive, and time-consuming tests, which may be impractical in certain situations. The Small Punch Test (SPT) is a small-sample detection technique for evaluating material mechanical properties, characterized by its "non-destructive" nature: applying localized loads to small specimen samples and measuring the resulting deformation. While SPT can obtain load-displacement curves for specimens, it cannot directly reflect material mechanical properties and relies on empirical formulas for estimation. To address this challenge, we designed a specialized SPT experiment and fixture, and established a Finite Element Method (FEM) model. We developed a multi-fidelity model capable of predicting the mechanical properties of steel and aluminum alloys, representing a novel machine learning

Zou, Jie, Li, Shanshan, Huayang, Xiang, Chen, Yechao

Image-based Machine Learning Methods in Materials Microstructure and Failure Analysis

2025-01-8324To be published on 04/01/2025

Image-based machine learning (ML) methods are increasingly transforming the field of materials science, offering powerful tools for automatic analysis of microstructures and failure mechanisms. This paper provides an overview of the latest advancements in ML techniques applied to materials microstructure and failure analysis, with a particular focus on the automatic detection of porosity and oxide defects and microstructure features such as dendritic arms and eutectic phase in aluminum casting. By leveraging image-based data, such as metallographic and fractographic images, ML models can identify patterns that are difficult to detect through conventional methods. The integration of convolutional neural networks (CNNs) and advanced image processing algorithms not only accelerates the analysis process but also improves accuracy by reducing subjectivity in interpretation. Key studies and applications are further reviewed to highlight the benefits, challenges, and future directions of

Akbari, Meysam, WANG, Andy, Wang, Qigui, Yan, Cuifen

Regression Model based Method for Stiffness Design of Laminated Glasses

2025-01-8320To be published on 04/01/2025

There has been a growing interest in laminated sidelites (or door glasses) in automotive for the benefits of both security and comfort. Current standard products are a 2.1/0.76/2.1 mm construction with soda-lime glass symmetrically distributed on each side of a PVB layer. Due to low shear stiffness of PVB, the standard laminate has a significantly reduced flexural rigidity in comparison to a monolith with the same total thickness. A novel lightweight laminate concept with ultrathin chemically strengthened glass was developed by Corning [1]. This concept considered asymmetrical designs with thinner (0.5 – 1.1 mm) chemically strengthened Corning® Gorilla® Glass laminated to relatively thick outer ply of conventional soda-lime glass and was found to provide enhanced stiffness over standard symmetrical laminate designs of same total thickness. It is known that flexural rigidity of laminated glass depends on multiple variables, such as shape, load distribution and boundary constraining

Yu, Chao

Utilizing Generative Adversarial Networks for Secure Communication in Software-Defined Vehicles

2025-01-8135To be published on 04/01/2025

The increasing complexity of software-defined vehicles (SDVs) necessitates robust and secure communication protocols to protect against cyber threats. This paper explores the utilization of Generative Adversarial Networks (GANs) to enhance the security of communication protocols in SDVs. GANs, consisting of a generator and a discriminator network, are employed to create and evaluate secure communication sequences, ensuring that unauthorized access and potential attacks are effectively mitigated. In this study, we develop a GAN-based framework that generates secure communication protocols tailored for the dynamic environment of SDVs. The generator is trained to produce communication sequences that are indistinguishable from authentic, secure sequences, while the discriminator is tasked with identifying any anomalies or potential vulnerabilities. By iteratively improving both networks, the framework learns to produce highly secure and resilient communication protocols. The performance of

Namburi, Venkata Lakshmi

Use of Vehicle Cyber Engineering (VCE) testbeds to develop AI Safety enhancements based on ISO TR 5469

2025-01-8090To be published on 04/01/2025

The University of Detroit-Mercy (UD-M) has developed CLaaS (Cyber-Security Labs as a Service) to support teaching students how to understand and mitigate cyber security attacks. The UD-M Vehicle Cyber Engineering (VCE) laboratory has physical hardware testbeds to support education and research on actual vehicle systems. With the increased use of Machine Learning (ML) and Artificial Intelligence (AI) in on- and offroad vehicles, it is becoming critical to enhance the laboratory to help educate a new generation of engineers that understand the cyber security and functional implications of AI/ML. The ISO TR 5469 Technical Report provides a framework to classify the AI/ML technology based on usage level and the properties and requirements to mitigate cyber and functional safety risks for the technology. This paper provides an overview of the approach used by ISO TR 5469 as well as an example how one of the six ISO TR 5469 desirable properties (resilience to adversarial and intentional

Zachos, Mark, Seifert, Heinz

Reproduction of Real-World Scenarios in CARLA: An Extension of CARLA Functionality

2025-01-8059To be published on 04/01/2025

Reproducing driving scenarios involving near-collisions and collisions in a simulator can be useful in the development and testing of autonomous vehicles, as it provides a safe environment to explore detailed vehicular behavior during these critical events. CARLA, an open-source driving simulator, has been widely used for driving scenario reproduction. However, current versions of CARLA are limited to setting the start and destination points for vehicles, and are unable to replay precise waypoint paths that are collected from real-world collision and near-collision scenarios, due to the fact that the collision-free pathfinding modules built into the system. This paper presents a significant extension to CARLA’s source code, enabling the replay of exact vehicle paths, irrespective of safety implications encountered during scenario layouts. With this functionality, researchers can now simulate real-world scenarios with higher fidelity, so that the vehicular automation can avoid unsafe

Ai, Yan, Adhikari, Bikram, Park, Chung-Kyu, Kan, Cing-Dao, Wijesekera, Duminda

Object Detection for City and Highway Driving Scenario with YOLOX and Mask RCNN

2025-01-8015To be published on 04/01/2025

This paper explores the integration of two deep learning models that are currently being used for object detection, specifically Mask R-CNN and YOLOX, for two distinct driving environments: urban cityscapes and highway settings. The hypothesis underlying this work is that different methods of object detection will work best in different driving environments, due to the differences in their unique strengths as well as the key differences in those driving environments. Some of these differences in the driving environment include varying traffic densities, diverse object classes, and differing scene complexities, including specific differences such as the types of signs present, the presence or absence of stoplights, and the limited-access nature of highways as compared to city streets. As part of this work, a scene classifier has also been developed to categorize the driving context into the two categories of highway and urban driving, in order to allow the overall object detection

Patel, Krunal, Peters, Diane

Lane-Keeping for SAE Level 3 Autonomous Vehicles: A Behavior-Tree Approach Using Lidar Data for Enhanced Safety and Control

2025-01-8025To be published on 04/01/2025

Lane-keeping is a critical function for SAE Level 3 autonomous vehicles, requiring rigorous validation and end-to-end interpretability. All recently U.S. approved level 3 vehicles are equipped with lidar, likely meant to accelerate active safety regulation compliance. Lidar offers precise and reliable surface distance and intensity measurements, enabling more effective implementation of rule-based algorithms compared to camera-based methods, which tend to rely on machine learning based methods for state-of-the-art performance. This paper presents a behavior-tree algorithm for lane-keeping using data from an Ouster OS0 lidar sensor. The algorithm evaluates factors such as road curvature, speed limits, road types (rural, urban, interstate), and the proximity of objects or humans to lane markings. It also accounts for the behavior and type of adjacent and opposing vehicles, lane occlusion, and weather conditions. The algorithm outputs safe lane keeping control instructions. The algorithm

Soloiu, Valentin, Mehrzed, Shaen, Kroeger, Luke, Pierce, Kody, Sutton, Timothy, Lange, Robin

Workload Estimation for a Military Ground Vehicle Crew using Supervised Machine Learning of FACS Action Unit Intensity Data

2025-01-8340To be published on 04/01/2025

The proliferation of intelligent technologies in the future battlefield necessitates an exploration of crew workload balancing strategies for human-machine integrated formations. Many current techniques to measure cognitive workload, through qualitative surveys or wearable sensors, are too brittle for the harsh, austere operational environments found in military settings. Non-invasive workload estimation techniques, such as those that analyze physiological effects from video feeds of the crew, present a way forward for workload-aware Soldier-machine interfaces that could trigger events – such as task reallocation – if limits on crew or individual workload are exceeded. One such technique that is being explored is the use of facial expression analysis for workload estimation. We present the performance results of regression and classification models developed from supervised machine learning algorithms that predict pNN50, a common heart rate variability metric used as a physiological

Mikulski, Christopher, Riegner, Kayla

A Study on Reconstructing In-Cylinder Combustion Images Based on Local Images

2025-01-8381To be published on 04/01/2025

The current leading experimental platform for engine visualization research is the optical engine, which features transparent window components classified into two types: partially visible windows and fully visible windows. Due to structural limitations, fully visible windows cannot be employed under certain complex or extreme operating conditions, leading to the acquisition of only local in-cylinder combustion images and resulting in information loss. This study introduces a method for reconstructing in-cylinder combustion images from local images using deep learning techniques. The experiments were conducted using an optical engine specifically designed for spark-ignition combustion modes, capturing in-cylinder flame images under various conditions with high-speed cameras. The primary focus was on reconstructing the flame edge, with in-cylinder combustion images categorized into three types: images where the flame edge is fully within the partially visible window, partly within the

Wang, Mianheng, Zhang, Yixiao, Du, Haoyu, Xiao, Ma, Mao, Jianshu, Fang, Yuwen

Optimization-Oriented Adaptive Equivalent Consumption Minimization Strategy for Fuel Cell Hybrid Electric Vehicles Based on Power

2025-01-8552To be published on 04/01/2025

Fuel economy and the ability to maintain the state of charge (SOC) of the battery are two key metrics for the energy management of a full-power fuel cell hybrid vehicle fitted with a small-capacity battery pack. To achieve stable maintenance of SOC and near-optimal fuel consumption, this paper proposes an adaptive equivalent consumption minimization strategy (PA-ECMS) based on power prediction. The strategy realizes demand power prediction through a hybrid deep learning model, and periodically updates the optimal equivalent factor (EF) based on the predicted power to achieve SOC convergence and ensure fuel economy. Simulation results show that the hybrid deep learning network model has high prediction accuracy with a root mean square error (RMSE) of only 0.733 m/s. Compared with the traditional ECMS based on SOC feedback, the PA-ECMS effectively maintains the battery SOC in a more reasonable range, reduces the situation of the fuel cell directly charging the power cell in the high

Gao, Xinyu, Ju, Fei, Chen, Gang, Zong, Yuhua, Wang, Liangmo

Fully Automated Global Powertrain Optimization for Electric Truck Using Active Learning Algorithm and Backwards Simulation Approach

2025-01-8536To be published on 04/01/2025

To reduce greenhouse gas emissions, electric vehicles (EVs) are increasingly becoming a significant part of the automotive market. This trend is not limited to passenger cars, but also extending to the truck segment. Electric trucks, due to their substantial self-weight and payload, require a different approach compared to electric passenger cars. While electric passenger cars generally use only one motor or a single-speed transmission, electric trucks necessitate the use of multiple motors or multi-speed transmissions. This addresses a complex issue: how to determine the size of the motor, the number of motors, the number of gears, and the corresponding gear ratios in the drive system of an electric truck. The nonlinearity of the vehicle system further complicates this issue. To tackle this, the paper proposes the application of an active learning algorithm, namely Bayesian optimization, to determine the configuration of the motors and transmission. The goal is to provide an optimal

Chen, Bicheng, Wellmann, Christoph, Xia, Feihong, Savelsberg, Rene, Andert, Jakob, Pischinger, Stefan

Machine Learning-Based Digital Twin for Optimizing Automotive Components

2025-01-8206To be published on 04/01/2025

This paper presents a Digital Twin approach based on Machine Learning (ML), aimed at creating software-based sensors to reduce the auxiliary devices of the vehicle and enabling predictive maintenance, thus reducing carbon footprint. The solution is applied to the Electric Lubrication Oil Pump (ELOP), a crucial component within a vehicle's powertrain system. The proposed ELOP Digital Twin integrates ML-based sensors to estimate critical parameters such as temperature, pressure and flow rate, reducing the reliance on physical sensors and associated hardware. This estimation approach minimizes manufacturing complexity and cost, enhancing energy efficiency during both production and operation. Furthermore, the digital twin facilitates predictive maintenance by continuously monitoring the component's performance, enabling early detection of potential failures and optimizing maintenance schedules. This leads to lower energy consumption and reduced emissions throughout the component's

Khan, Jalal, D'Alessandro, Stefano, Tramaglia, Federico, Fauda, Alessandro

Cloud Based Digital Twin Development for High-Voltage Battery Systems in Commercial Vehicles

2025-01-8214To be published on 04/01/2025

Over recent years, BorgWarner has intensified its efforts to explore and leverage trending technologies such as Artificial Intelligence (AI) and Machine Learning (ML) to enhance products and processes. The digital twin technology has potential use cases on system behavior analysis, product optimization and predictive maintenance. As the high-voltage battery is the most valuable component in any electric vehicle understanding the mechanisms of battery aging and failure, along with their dependencies, is crucial for improving performance, battery lifespan, and residual value. A digital twin not only aids in comprehending the impacts of user behavior on the battery's state of health but also facilitates further fault diagnosis and analysis. This paper outlines the development process of a digital twin for a commercial vehicle battery. A cloud-based digital twin demonstrator was developed, which integrates vehicle telemetry data with physics-based battery electric and thermal models, and

Bongards, Anita, Liu, Xiaobing, Beemer, Maria, Gajowski, Daniel, Rama, Neeraj, Shah, Keya, Fallahdizcheh, Amirhossein

Application of Machine Learning Model on Automotive Subframe Design

2025-01-8619To be published on 04/01/2025

The automotive subframe, also referred to as a cradle, is a critical chassis structure that supports the engine/electric motor, transmission system, and suspension components. The design of a subframe requires specialized expertise and a thorough evaluation of performance, vehicle integration, mass, and manufacturability. Suspension attachments on the subframe are integral, linking the subframe to the wheels via suspension links, thus demanding high performance standards. The complexity of subframe design constraints presents considerable challenges in developing optimal concepts within compressed timelines. With the automotive industry shifting towards electric vehicles, development cycles have shortened significantly, necessitating the exploration of innovative methods to accelerate the design process. Consequently, AI-driven design tools have gained traction. This study introduces a novel AI model capable of swiftly redesigning subframe concepts based on user-defined raw concepts

Yang, Jiongzhi, Sarkaria, Bikramjit, Kumaraswamy, Prashanth, Kailkere Srinivas, Praveen

Identifying Negative Driver States that Share Commonalities for Interventions

2025-01-8676To be published on 04/01/2025

Advancements in sensor technologies have led to increased interest in detecting and diagnosing "driver states"-collections of internal driver factors generally associated with negative driving performance, such as alcohol intoxication, cognitive load, stress, and fatigue. This is accomplished using imperfect behavioral and physiological indicators that are associated with those states. An example is the use of elevated heart rate variability, detected by a steering wheel sensor, as an indicator of frustration. Advances in sensor technologies, coupled with improvements in machine learning, have led to an increase in this research. However, a limitation is that it often excludes naturalistic driving environments, which may have conditions that affect detection. For example, reductions in visual scanning are often associated with cognitive load (e.g., Angell et al., 2015); however, these reductions can also be related to novice driver inexperience (e.g., Stahl et al., 2019) and alcohol

Seaman, Sean, Zhong, Peihan, Angell, Linda, Domeyer, Joshua, Lenneman, John

Development of deep reinforcement learning traction controllers for front and rear wheel drive electrified vehicles

2025-01-8803To be published on 04/01/2025

Traction control plays a key role in improving vehicle safety, especially for driving scenarios involving low levels of tyre-road friction. Over the past 30 years, academic and industrial research in traction controllers has mainly favoured deterministic approaches. This paper introduces a traction control strategy based on a deep reinforcement learning agent tailored for straight-line acceleration manoeuvres from standstill in low-friction conditions. The proposed agent is trained on two different electric vehicles, a front-wheel drive city car (from EU vehicle segment A), and a rear-wheel drive sedan (from EU vehicle segment D). The paper presents a deep reinforcement learning agent formulation suitable for training on different vehicles, assesses the performance of the resulting controllers, and evaluates their response to changes in vehicle mass, powertrain parameters and tyre-road friction conditions. The assessment uses a high-fidelity co-simulation model, combining AVL VSM and

Caponio, Carmine, Mihalkov, Mario, Hankovszki, Zoltan, Fuse, Hiroyuki, Ivanov, Valentin, Sorniotti, Aldo, Gruber, Patrick, Montanaro, Umberto

Introducing the ML FMEA: The Next Step in Machine Learning Safety

2025-01-8078To be published on 04/01/2025

Several challenges remain in deploying Machine Learning (ML) into safety critical applications. We introduce a safe machine learning approach tailored for safety-critical industries including automotive, autonomous vehicles, medical automation, industrial robotics, and aerospace. Aiming to fill a perceived gap within Artificial Intelligence and ML standards, the described approach integrates ML best practices with the proven Process Failure Mode & Effects Analysis (PFMEA) approach to create a robust ML pipeline. The solution views ML development holistically as a value-add, feedback process rather than the resulting model itself. By applying PFMEA, the approach systematically identifies, prioritizes, and mitigates risks throughout the ML development pipeline. The paper outlines each step of a typical pipeline, highlighting potential failure points and tailoring known best practices to minimize identified risks. As an additional contribution, a populated ML FMEA template is provided

Schmitt, Paul, Pennar, Krzysztof, Lopez, Jerry, Bijelic, Mario, Seifert, Heinz Bodo

Control Trajectory Optimization of Electric Vehicle Heat Pump-based Cabin Heating System

2025-01-8144To be published on 04/01/2025

Optimal control of battery electric vehicle thermal management systems is essential for maximizing the driving range in extreme weather conditions. Vehicles equipped with advanced heating, ventilation and air-conditioning (HVAC) systems based on heat pumps with secondary coolant loops are more challenging to control due to actuator redundancy and increased thermal inertia. This paper presents the dynamic programming (DP)-based offline control trajectory optimization of heat pump-based HVAC aimed at maximizing thermal comfort and energy efficiency. Besides deriving benchmark results, the goal of trajectory optimization is to gain insights for practical hierarchical control strategy modifications to further improve real-time controllers’ performance. DP optimizes cabin inlet air temperature and flow rate to set the trade-off between thermal comfort and energy efficiency while considering the nonlinear dynamics and operating limits of HVAC system in addition to typically considered cabin

Cvok, Ivan, Deur, Josko

Machine Learning-Enabled Optimization of Vehicle Restraint Systems – Demonstration in a Real-world Crash Scenario

2025-01-8722To be published on 04/01/2025

Vehicle restraint systems, such as seat belts and airbags, play a crucial role in crash energy management and occupant protection during vehicle crash events. Designing an effective restraint system for a diverse population is a highly complex task, as it requires a seamless integration of essential components. This study demonstrates the practical implementation of state-of-the-art Machine Learning (ML) techniques to optimize vehicle restraint systems and improve occupant safety. A ML-based surrogate model was developed using a limited Design of Experiments (DOE) dataset from finite element human body model simulations and employed to optimize the vehicle restraint system. The performance of the ML-optimized restraint system was compared with the baseline design in a real-world crash scenario. Despite the baseline design having already been optimized during the vehicle development phase, the ML-based optimization resulted in additional improvements. Specifically, in the US NCAP

Lalwala, Mitesh, Lin, Chin-Hsu, Desai, Megha, Rao, Shishir

Parameter Virtual Calibration Method for Vehicle Path Tracking Controller Based on the Deep Reinforcement Learning

2025-01-8311To be published on 04/01/2025

Path tracking control, which is one of the most important foundations of autonomous driving, could help the vehicle to precisely and smoothly follow the preset path by actively adjusting the front wheel steering angle. Although there are a number of advanced control methods with simple structure and reliable robustness that could assist vehicles achieving path tracking, these controllers have many parameters to be calibrated, and there is a lack of guidance documents to help non-professional test site engineers quickly master calibration methods. Therefore, this paper proposes a parameter virtual calibration method based on the deep reinforcement learning, which provides an effective solution for parameter calibration of vehicle path tracking controller. Firstly, the vehicle trajectory tracking model is established through the kinematic relationship between the vehicle and the target path, combined with the Taylor series expansion linearization method. Next, a vehicle path tracking

Zhao, Jian, Guo, Chenghao, Zhao, HuiChao, Zhao, Yongqiang, Yu, Zhen, Zhu, Bing, Chen, Zhicheng

Research on Turbocharger Surge Identification and Prediction Based on Acoustic Signal

2025-01-8253To be published on 04/01/2025

This study presents a non-intrusive method for detecting surge in diesel engine turbochargers using acoustic signal analysis. We introduce an innovative surge identification approach that leverages multi-domain composite features derived from these acoustic signals. The research involved conducting controlled turbocharger surge experiments, which provided the data necessary to develop a comprehensive surge acoustic signal database. From this database, eighteen acoustic features were meticulously selected and extracted across both time and frequency domains. Their relevance to surge detection was evaluated using a random forest algorithm, enabling the identification of key features critical to accurate detection. These essential features were then integrated into a multi-domain composite feature set, which served as the sensitive indicator for turbocharger surge detection. During the model development phase, we systematically compared the performance of various machine learning

zhu, jiaxu, Zheng, Hongyu, Zong, Changfu

Development of Prediction Model for Vehicle Road Load Using Machine Learning

2025-01-8258To be published on 04/01/2025

In the modern-day automotive industry, improving fuel efficiency while reducing carbon emissions is a critical challenge. To meet this challenge, it is essential to accurately measure the road load of a vehicle. The current methodology, which is widely adopted by major national guidelines, is to follow the coastdown test procedure. However, coastdown tests are highly influenced by environmental conditions, making it difficult to obtain consistent results for each test run. Previous studies mainly focused on the effect of independent variables on the coastdown results and not from a data-driven perspective due to the difficulty in obtaining a massive amount of test data in a short period of time, both in terms of time and cost. This paper presents a road load energy prediction model for vehicles using the XGBoost machine learning technique and how it can be used to predict the road load coefficients. The model features 27 factors, including rolling, aerodynamic, inertial resistance, and

Song, Hyunseung, LEE, DONG HYUK, Chung, Hyun

Uncertainty Quantification in Machine Learning using an Ensemble Approach with Gaussian Process Regression

2025-01-8199To be published on 04/01/2025

Machine learning has witnessed widespread adoption across various domains, bringing about transformative changes in decision-making, trend prediction, task automation, and personalized experiences. Despite the remarkable predictive capabilities of machine learning models, the associated uncertainty in their predictions remains a critical concern. Uncertainty estimation plays a pivotal role in ensuring robust decision-making, going beyond mere outcome prediction to quantify the model's confidence and potential error. This paper first presents a review of existing uncertainty quantification techniques in machine learning, including Monte Carlo dropout and ensemble methods, highlighting their advantages in addressing uncertainty as well as their limitations. Then, it presents an efficient and fast novel technique for uncertainty quantification using a combination of the ensemble technique and Gaussian process regression providing an accurate estimation of uncertainty bounds. Due to its

Chavare, Sudeep, Mourelatos, Zissimos P.

Optimization of Electric Vehicle Battery Charging Strategy Based on Reinforcement Learning

2025-01-8116To be published on 04/01/2025

The battery capacity of electric vehicles (EVs) is a critical factor influencing their overall performance. Selecting an appropriate charging strategy based on the EV driver’s travel behaviors can significantly mitigate battery capacity degradation and prolong its lifespan. This study integrated a physics-based battery degradation model with an EV powertrain system energy consumption model to enhance the prediction accuracy for the battery status under real-world driving conditions. In addition, it employed the Q-learning algorithm to provide suggestions on the driver’s charging behaviors, including charging start time, charging duration, and charger types. To create a reasonable reward function in the Q-learning algorithm, this study addressed the issue of car rental costs when the battery capacity is insufficient to support daily mileage. With the algorithm, this study modeled the 3-year use of an EV by an average driver sampled from the New England area in the U.S. The simulation

Wang, Jiayi, Jing, Hao, Ou, Shiqi(Shawn), Lin, Zhenhong

Assessment of Injuries in Rear Impacts

2025-01-8727To be published on 04/01/2025

Rear impacts make up a significant portion of crashes in the United States. To date, regulations on rear impacts have focused on fuel system integrity and seat performance, while most research has focused on seat performance in relation to occupants’ injuries, with some analyses of crash severity and seat belt effects. The performance of seats and seat belts may vary depending on the size of the occupant. Understanding how occupant characteristics, as well as crash scenarios, affect injury outcomes can show opportunities for further enhancements in rear impact occupant protection. This paper presents analyses using survey weighted logistic regression models to understand the factors affecting serious injury outcomes (i.e., MAIS 3+) in rear impacts, exploring the potential for improving occupant outcomes. Three separate models are evaluated, focusing on 1) overall injury level, 2) head, neck and cervical-spine injuries, and 3) thorax, abdomen, thoracic- and lumbar-spine injuries for

Greib, Joshua, jurkiw, Renee, Kryzaniwskyj, Tanja, Owen, Susan, Van Rooyen, Paul, Whelan, Stacey, Williamson, John

Artificial Intelligence Approach for Fuel Cell Model Parameter Calibration

2025-01-8632To be published on 04/01/2025

Performance of fuel cells can be characterized by a polarization curve which represents the current-voltage dynamics of the fuel cell. Such a polarization curve can be generated through a Computational Fluid Dynamics (CFD) simulation by running a series of simulations at various load voltages to calculate the current produced. The electrochemical model used in such a CFD simulation involves many parameters. For a numerical CFD fuel cell model to match experimental polarization curve, it is critical to find the right values of these parameters. It is hard to find the values of all the parameters experimentally, and hence parameter calibration is required. Traditional calibration methods, which often involve manual trial-and-error, can be time-consuming and prone to uncertainty. This work introduces an Artificial Intelligence (AI) assisted parameter calibration framework designed to enhance the efficiency and accuracy of model tuning. By integrating machine learning techniques

Champhekar, Omkar, Janakiraman, Arun, Gondipalle, Sreekanth, Ajotikar, Nikhil, Zehr, Randall

ROADSIDE TRAFFIC SURVEILLANCE FOR OPTIMAL TRAFFIC CONTROL USING A REAL-TIME EDGE COMPUTING PLATFORM

2025-01-8099To be published on 04/01/2025

In recent years, traffic surveillance has become a crucial component of Intelligent Transportation Systems (ITS), playing a significant role in enhancing transportation safety, improving transit efficiency, refining traffic management and control, and fostering sustainability in urban and rural environments. The recent advancements in machine learning and sensor fusion technologies are expected to further increase the importance of traffic surveillance within ITS. Although traffic surveillance can be conducted through either centralized or distributed network architectures, the latter, enabled by edge computing, is becoming increasingly popular. Consequently, this research concentrates on leveraging deep learning and sensor fusion technologies for traffic surveillance on an edge computing platform. The technical objective is to develop a comprehensive system capable of detecting, classifying, localizing, and tracking vehicles using roadside sensors and AI computing hardware. For

Li, Shenglin, Leech, David, Yoon, Hwan-Sik

A deep learning framework for time series prediction of passenger motion sickness based on vehicle dynamics data

2025-01-8104To be published on 04/01/2025

Technology development for enhancing passenger experience has been gaining attention as autonomous vehicles (AVs) provide new opportunities for human-vehicle interaction. A new possibility for occupants of AVs is performing productive tasks as they’re relieved from the task of driving. However, it has been shown that non-driving-related tasks cause a conflict between the visual and vestibular sensory systems and can, consequently, lead to Motion Sickness (MS). To develop MS mitigating technologies, it’s advantageous to design a tool that can predict when MS is likely to occur. However, there is currently a lack of in-vehicle systems that are capable of estimating a passenger's MS state in real-time. Furthermore, the lack of real-time physiological data from vehicle occupants constrains the types of sensory data that can be used for estimation under realistic implementations. To address this, a computational model was developed to predict the MS score for passengers solely based on the

Kolachalama, Srikanth, Sousa Schulman, Daniel, Kerr, Bradley, YIN, Siyuan, Wachsman, Michael Ben, Pienkny, Jedidiah Ethan Shapiro, Jalgaonkar, Nishant M., Awtar, Shorya

Developing a Vehicle-in-Virtual-Environment (VVE) Based Autonomous Driving Function Development and Evaluation Pipeline for Vulnerable Road User Safety

2025-01-8061To be published on 04/01/2025

Even though research in autonomous driving is progressing rapidly, with many successful deployments and the expectation of significantly reducing car accidents caused by human error, the number of injuries and fatalities involving vulnerable road users has kept increasing. Traditional methods for developing and evaluating autonomous driving functions, such as model-in-the-loop (MIL) and hardware-in-the-loop (HIL) simulations, heavily depend on the accuracy of simulated vehicle models and human factors, especially for vulnerable road user safety systems. Continuation of development during public road deployment forces other road like including vulnerable ones to involuntarily participate in the development process, leading to safety risks, inefficiencies, and a decline in public trust. To address these deficiencies, the Vehicle-in-Virtual-Environment (VVE) method was proposed as a safer, more efficient, and cost-effective solution for developing and testing connected and autonomous

Chen, Haochong, Cao, Xincheng, Guvenc, Levent, Aksun Guvenc, Bilin

Combine HD map information to enhance road topology reasoning by offline supervised training and on-line ensemble learning

2025-01-8021To be published on 04/01/2025

Topology reasoning plays a crucial role in understanding complex driving scenarios and facilitating downstream planning, yet the process of perception is inevitably affected by weather, traffic obstacles and worn lane markings on road surface. Combine pre-produced high-precision maps, and other type of map information to the perception network can effectively enhance perception robustness, but this on-line fused information often requires a real-time connection to website servers. We are exploring the possibility to compress the information of offline maps into a network model and integrate it with the existing perception model. We designed a topology prediction module based on graph attention neural network and an information fusion module based on ensemble learning. The module, which was pre-trained on offline high-precision map data, when used online, inputs the structured road element information output by the existing perception module to output the road topology, and the output

kuang, Quanyu, rui, Zhang, Zhang, Song, yixuan, gao

Neural Network Based Kalman Filter Design for a Vehicle Lateral Maneuver

2025-01-8042To be published on 04/01/2025

Precise state estimation during a lateral maneuver is not just a theoretical concept, but a practical necessity. The performance of Kalman filter is directly impacted by the comprehensive research and innovative approaches to counter nonlinearity and uncertainty. The use of machine learning in control theory is one such development that has significantly enhanced the effectiveness of our work. This paper provides an enhanced adaptive Kalman filter architecture with neural network for a rapid obstacle avoidance maneuver. The proposed design exemplifies its effectiveness in terms better state estimation in the presence of complex nonlinear vehicle dynamics and disturbances. Simulation results verify the same by ensuring a significant improvement to the traditional design, demonstrating better accuracy and the need for such advances in vehicle dynamics and control

Sudhakhar, Monish Dev

Role of AI in an Autonomous Vehicle Perception Systems: Use of Convolutional Neural Network Approach to Design the Vehicle Perception System

2025-01-8012To be published on 04/01/2025

Artificial Intelligence has gained lot of traction and importance in the 21st century with use cases ranging from Speech recognition, Learning, Planning, Problem solving to search engines etc. Artificial Intelligence also has played a key role in the development of Autonomous Vehicles and robots ranging from perception, localization, decision to controls. Within the big AI umbrella there is Machine learning which is all about using your computer to "learn" how to deal with problems without “programming". Deep learning is a branch of machine learning based on a set of algorithms that learn to represent the data directly from the input such as an Image, text, Sound, etc. Within deep learning there are Convolutional Neural Networks and Recurrent Neural Networks (CNN/RNN). The study here used convolutional neural network approach to perform image/object recognition. Given that the objective of the autonomous or semi-autonomous vehicle is to promote safety and reduce number of accidents, it

Mansour, Iyad, Singh, Sanjay

Machine Learning Based Variation Studies for Electric Vehicle Drive Unit Virtual System Models

2025-28-0171To be published on 02/07/2025

In electric vehicles development, manufacturing variations pose big challenge in designing various mechanical components as these variations directly impact various customer perceivable performance outputs. If the manufacturing variations can be included in design phase itself, overall robustness of the design can be enhanced. This paper delineates machine learning based methods to include manufacturing variations in designing drive units for electric vehicles. In an electric vehicle, the drive unit transfers power or torque from a battery through an inverter to wheels. The drive units are subjected to different types of loads under various vehicle maneuvers. To evaluate the drive unit system virtually, system level simulations are performed. Traditionally, nominal values of the several inputs such as bearing parameters, gear parameters and clearances etc. are used. However, the drive unit must be designed in such a way that outputs meet target considering all the variations of inputs

Penumatsa, Venkata Ramana Raju, Katha, Venkateswar, Black, Derrick, Jain, Sachin, Maddipati, Seshagiri

Gear misalignment optimization for Electric Drive Unit using Machine learning based Prescriptive Analytics

2025-28-0166To be published on 02/07/2025

Gear mesh misalignment in electric drive unit may result in shifts in the load distribution of a gear pair that can increase contact and bending stresses. It may move the peak bending and contact stresses to the edge of the face width, and also increase gear noise. Lower misalignment value is often required to reduce the peak bending and contact stresses and have a balanced load distribution along the gear flank. This paper delineates a method which is also called prescriptive analytics that combines virtual simulations, Machine learning (ML) and optimization techniques to minimize different gear misalignments for electric vehicle drive unit. The gear misalignments are dependent on stiffness of different components in the drive unit. So, a comprehensive approach that incorporates effect of individual components and their interactions is needed. Generally, an assembly level virtual simulation is performed to evaluate gear misalignments. In order to minimize the misalignments, manual

Penumatsa, Venkata Ramana Raju, Thomas, Benson, Black, Derrick, Jain, Sachin

Enhancing O-Ring Performance Prediction Using Machine Learning

2025-28-0161To be published on 02/07/2025

O-rings are essential sealants in engineering applications, ensuring leak-tight seals, and preventing fluid leakage. Specifically, in Fuel Cells, they play a crucial role in maintaining system integrity by preventing working fluid leakages. Finite Element Analysis (FEA) is commonly employed for virtual assessments of sealant assemblies. By simulating the behavior of O-rings under various conditions, engineers can evaluate their performance. However, the traditional approach of varying individual input parameters to understand their impact on performance is time-consuming and extensive. To address this challenge, machine learning (ML) algorithms offer an efficient alternative to repetitive traditional FEA. By leveraging ML models, engineers can predict output parameters more effectively, streamlining the analysis process and enhancing accuracy. Despite the benefits of ML, different predictive models exhibit varying prediction performance, leading to inconsistency. Therefore, it is

Mallu, Venkata Reddy, Penumatsa, Venkata Ramana Raju, Chirravuri, Bhaskara, Duddu, Varaprasad, Miller, Ronald, Sahu, Abhishek

A Comparative Study on fault diagnosis of Electrical vehicle motor testing machine Gear Components Using machine Learning algorithms

2025-28-0177To be published on 02/07/2025

Electric vehicles (EVs) are the way of the future for the automobile industry. The drive motor is a vital part of modern EVs. A motor testing machine is used to verify the critical parameters of the electrical motor, which is seen to be of the utmost importance. Electric motor characteristic curve points and their electromagnetic behaviour under different conditions are regularly found using the motor testing machine. But it has been noted that a flaw in the testing machine's helical gear arrangement causes frequent malfunctions and less-than-ideal performance, which impacts the assembly line's effectiveness in producing electric cars. The research project is to enhance the motor test bench apparatus by modifying crucial design parameters using machine learning and vibration signal analysis methods. Vibration signals are recorded at various gear settings before statistical features are extracted. Following that, these signals are classified using classifiers such Quadratic SVM and

S, Ravikumar, Syed, Shaul, V, Muralidharan, D, Pradeep Kumar

Autonomous Drone Solution for Human-Wildlife Conflict Management

2025-28-0198To be published on 02/07/2025

Human-wildlife conflicts pose significant challenges to both conservation efforts and community well-being. As these conflicts escalate globally, innovative technologies become imperative for effective and humane management strategies. This paper presents an integrated autonomous drone solution designed to mitigate human-wildlife conflicts by leveraging technologies in drone surveillance and artificial intelligence. The proposed system consists of stationary IR cameras that are setup within the conflict prone areas, which utilizes machine learning to identify the presence of wild animals and to send the corresponding location to a drone docking station. An autonomous drone equipped with high resolution IR cameras and sensors is deployed from the docking station to the provided location. The drone camera utilizes object detection technology to scan the specified zone to detect the animal and emit animal repelling ultrasonic sound from a device integrated to the drone to achieve non

Sadanandan, Vaishnav, Sadique, Anwar, George, Angeo Pradeep, Vinod, Vishal, Raveendran, Darshan Unni

Items per page:

1 – 50 of 1059