Introducing machine learning (ML) into safety-critical systems presents a fundamental challenge, as traditional safety analysis techniques often struggle to capture the dynamic, data-driven, and non-deterministic behavior of learning-enabled components. To address this gap, the Machine Learning Failure Mode and Effects Analysis (ML FMEA) methodology was developed as an open-source framework tailored to ML-specific risks. This paper reports on the maturation of ML FMEA from an initial conceptual framework to a proven, practice-driven methodology. We make four primary contributions. First, we extend the ML FMEA pipeline with two new stages: a “Step Zero” for problem definition and system-level hazard analysis, and a “Step 5” for constructing ground truth or reward signals. Autonomous vehicle and humanoid robot applications are presented to illustrate the practical application and safety benefits of these additions. Second, we introduce tailored Severity, Occurrence, and Detection

Schmitt, Paul, Shinde, Chaitanya, Diemert, Simon, Pennar, Krzysztof, Seifert, Bodo, Poh, Justin, Lopez, Jerry, Mannan, Fahim, Mohammed, Majed, Chalana, Akshay, Wadhvana, Neil, Wagner, Michael

Prediction of Battery Self-Discharge Under Sparse Data Using Physics-Guided Machine Learning

2026-01-0155

4/7/2026

Predicting battery self-discharge across wide temperature ranges and extended durations remains a significant challenge due to the scarcity of physical test data, which is typically limited to a few temperature points and short observation windows. This limitation complicates generalization and increases the risk of inaccurate extrapolation. To address this, the paper introduces a machine learning–based framework designed to predict self-discharge behavior under diverse thermal conditions and longtime horizons. Multiple modeling strategies are examined, including feedforward neural networks, long short-term memory (LSTM) architectures, synthetic data generation, and physics-informed integration of governing equations. Particular emphasis is placed on hybrid and physics-regularized models that embed first-principles relationships to guide extrapolation beyond the observed data domain. This approach mitigates the inherent instability and potential errors associated with purely data

Chavare, Sudeep, Zeng, Yangbing, Muppana, Sai Siddhartha, Miao, Yong, Xu, Simon

Improving EV Energy Forecast Accuracy via ML Residual Correction: The Role of Wind Speed in Data-Driven Predictions

2026-01-0150

4/7/2026

The increasing demand for electrified transportation is leading to accelerated development of highly efficient hybrid and battery electric vehicles. A major concern for customers adapting to battery electric vehicles (BEV) is range anxiety due to low charging speeds, charging infrastructure not matching expectations and unreliable range estimations shown to the customers by their vehicles. Estimating the range more accurately has been difficult due to the sensitivity of vehicle’s energy consumption to real-world environmental and driving conditions. This paper aims to find out the effect of true wind in the road load experienced by BEVs in the real-world driving scenarios and how using a highly accurate wind speed measurement improves the energy consumption estimation better. On-road tests were conducted on public roads and in controlled test-track environments to collect reliable wind speed measurements using a dynamic multi-hole pressure probe. Additional coastdown tests were also

Raghupathy, Vishnu Prasaad, Kim, Shinhoon, Evans, Nic, Niimi, Keisuke, Mochihara, Takahiro

A Dataset for Visual Classification of Flat Tires

2026-01-0152

4/7/2026

Flat tires represent a common yet serious issue in vehicle safety, leading to compromised control, increased braking distance, and potential rim or structural damage when undetected. Conventional tire pressure monitoring systems (TPMS) rely on embedded sensors that can fail, incur high replacement costs, and are not always equipped in older or low-cost vehicles. To address these limitations, this study presents a comprehensive visual dataset for flat-tire classification using computer vision and machine learning techniques. The dataset comprises 600 labeled images—300 flat-tire and 300 non-flat-tire samples—collected from diverse vehicle types, lighting conditions, and viewpoints. This dataset is designed to support the training and benchmarking of lightweight edge-AI models suitable for real-time deployment on embedded platforms. A set of supervised learning models were evaluated. Results demonstrate that visual-based classification provides a cost-effective and scalable pathway

Gunasekaran, Aswin, Govilesh, Vidarshana, Challa, Karthikeya, Maxim, Bruce, Shen, Jie

Transformer-Based Deep Learning Framework for Efficient Prediction of Automotive Wind Noise

2026-01-0222

4/7/2026

Aerodynamic wind noise is a critical challenge in modern automotive development, particularly with the rise of vehicle electrification and intelligent mobility, where cabin acoustic comfort is a key quality metric. While reliable, traditional methods like wind tunnel experiments and computational fluid dynamics (CFD) simulations are both costly and time-consuming. To address these challenges, we propose a novel Transformer-based framework for rapid and accurate wind noise prediction. Several model improvements, including the physical attention, geometry wave number embedding, hybrid FPS-random downsampling method and frequency separation output heads are properly employed to reduce the GPU memory cost and improve the prediction accuracy. This framework is pre-trained on a large-scale acoustic dataset of nearly 1,000 diverse vehicles generated using Improved Delayed Detached Eddy Simulation (IDDES). From a vehicle's point cloud coordinates, the model directly predicts the surface

Tang, Weishao, Liu, Mengxin, Qin, Ling, Duan, Menghua, Wang, Chengjun, Zhang, Yufei, Wang, Qingyang

Machine Learning-Based Prediction of Wind-Induced Interior Noise in Ground Vehicles

2026-01-0599

4/7/2026

In response to increasing customer demand for enhanced passenger comfort and perceived vehicle quality, OEMs in automotive and commercial vehicles are placing significant emphasis on reducing the interior cabin noise. At highway speeds, wind noise is a primary contributor to the overall noise within the vehicle cabin. Conventional approaches to predict vehicle wind noise rely on physical testing, which can only be conducted in the later stages of the design process once a physical prototype is available. Increased adoption of established computational fluid dynamics (CFD) methods has enabled earlier assessment. However, such simulations require several hours to complete, posing a challenge in the context of rapid design iteration cycles. With the growing adoption of artificial intelligence in engineering, machine learning (ML) approaches have been proposed to predict a vehicle’s aerodynamics performance. Nevertheless, development of ML techniques in the context of aeroacoustics

Higgins, John, Fougere, Nicolas, Sondak, David, Senthooran, Sivapalan, Moron, Philippe, Jantzen, Andreas, Bi, Jing, Oancea, Victor

Multi-Physics Topology Optimization and Neural Network–Based Design of Experiments for Lightweight Automotive Cast Components

2026-01-0510

4/7/2026

This paper presents a hybrid optimization framework that integrates Multi-Physics Topology Optimization (MPTO) with a Neural Network–surrogated Design of Experiments (NN-DOE) to enable lightweight structural design while satisfying crashworthiness, durability, and noise, vibration, and harshness (NVH) requirements under practical casting and packaging constraints. In the proposed MPTO formulation, crash and durability performances are incorporated through equivalent static compliance measures, while NVH performance is assessed using a frequency-domain dynamic stiffness metric, allowing consistent evaluation of trade-offs among competing design requirements. The framework is first demonstrated using a mass-produced passenger-car lower control arm (LCA) as a benchmark component. In this application, MPTO achieves weight reduction under multi-physics objectives by removing non-load-bearing material. Results show that single-discipline optimization produces unbalanced topologies, while

Kim, Hyosig, Senkowski, Andres, Gona, Kiran, Saroha, Lalit, Boraiah, Mahesh

A Latent Space Framework for Modeling Transient Engine Emissions Using Joint Embedding Predictive Architectures

2026-01-0423

4/7/2026

Accurately modeling and controlling vehicle exhaust emissions, particularly during highly transient events such as rapid acceleration, is crucial for meeting stringent environmental regulations and optimizing modern powertrain systems. While conventional data-driven modeling methods, such as Multilayer Perceptrons (MLPs) and Long Short-Term Memory (LSTM) networks, have improved upon earlier phenomenological or physics-based models, they often struggle to capture the complex nonlinear dynamics of emission formation. These monolithic architectures attempt to learn from all available data, which increases their sensitivity to dataset variability. They often require increasingly deep and complex architectures to improve performance, thereby limiting their practical utility. This paper introduces a novel approach that overcomes these limitations by modeling emission dynamics in a structured latent space. Using a rich dataset combining real-world driving data from a Portable Emission

Sundaram, Ganesh, Gehra, Tobias, Ulmen, Jonas, Heubaum, Mirjan, Görges, Daniel, Günthner, Michael

Computational Modeling and Optimization of Shape Memory Polymer-Based Energy Absorbers

2026-01-0573

4/7/2026

Shape memory polymers (SMPs) provide tunable thermomechanical properties and enable the design of recoverable crash structures for automotive applications. This paper introduces a computational framework for the design and optimization of SMP-based crash absorbers with periodic auxetic microstructures. First, a finite element (FE) model is developed and validated against experimental data regarding crushing and recovery behavior. A parametric study is then performed by varying key microstructural features, including wall thickness, cell size, and cell shape. Structural performance is evaluated in terms of specific energy absorption (SEA), peak force, and recoverability. To efficiently explore the high-dimensional design space, surrogate models based on machine learning are constructed, and multi-objective optimization is carried out to identify Pareto-optimal designs that balance competing objectives. The parametric study indicated that geometric parameters strongly influenced energy

Zhu, Yingbo, Zhu, Feng, Deb, Anindya

Adaptive Eco-Driving Feedback for Battery Electric Vehicles: A Reinforcement Learning Approach Using Context-Aware Policy Optimization

2026-01-0165

4/7/2026

Energy efficiency and range optimization remain critical challenges to the widespread adoption of battery electric vehicles (BEVs). As a result, there is a growing demand for intelligent driver assistance systems that can extend the operating range and reduce range anxiety. This paper presents an adaptive eco-feedback and driver rating system based on proximal policy optimization (PPO) reinforcement learning, designed to support drivers with the target to reduce energy consumption and maximize driving range. The system processes real-time driving data, such as velocity, acceleration and powertrain status. Map data of high quality is used to anticipate traffic events, including but not limited to speed limits, curves, gradients, preceding vehicles and traffic lights. This contextual awareness allows the system to continuously assess driving behavior and provide personalized, context-aware visual feedback alongside a dynamic driving behavior rating. A PPO agent learns optimal feedback

Stocker, Christoph, Hirz, Mario, Martin, Michael, Kreis, Alexander, Stadler, Severin

Machine Learning-Based Gear Defect Detection in Electric Vehicle Drive Units Using End-of-Line Vibration Analysis

2026-01-0637

4/7/2026

The final assembly of electric vehicle (EV) drive units includes an essential End-of-Line (EOL) test to ensure both component integrity and Noise, Vibration, and Harshness (NVH) quality. This screening process, which uses dynamometers to measure vibration signals, is critical for identifying defects before a drive unit is installed in a vehicle. A significant source of failure during this test is gear defects, which can arise from manufacturing or handling issues. Traditional EOL testing methods rely on time-domain analysis and the impulsiveness of vibration signatures to detect these defects, a technique with inherent limitations in accuracy. This paper introduces and evaluates a novel approach using Machine Learning (ML) to analyze vibration signals for improved gear defect detection. We discuss the methodologies of both the traditional time-domain and the proposed ML-based techniques. Finally, we provide a comprehensive comparison of their respective efficiency and accuracy

Arvanitis, Anastasios, Michaloliakos, Anargyros

Automotive Crash Dynamics Modeling Accelerated with Machine Learning

2026-01-0568

4/7/2026

Crashworthiness assessment is a critical aspect of automotive design, traditionally relying on high-fidelity finite element (FE) simulations that are computationally expensive and time-consuming. This work presents an exploratory comparative study on developing machine learning-based surrogate models for efficient prediction of structural deformation in crash scenarios using the NVIDIA PhysicsNeMo framework. Given the limited prior work applying machine learning to structural crash dynamics, the primary contribution lies in demonstrating the feasibility and engineering utility of the various modeling approaches explored in this work. We investigate two state-of-the-art neural network architectures for modeling crash dynamics: MeshGraphNet, a graph neural network that is widely employed in physics-based simulations, and Transolver, a transformer-based architecture with a physics-aware attention mechanism designed to maintain linear computational complexity with respect to geometric

Nabian, Mohammad Amin, Chavare, Sudeep, Akhare, Deepak, Ranade, Rishikesh, Cherukuri, Ram, Tadepalli, Srinivas

Design and Evaluation of a Deep Neural Network Model for Prediction of Diesel Equilibrium Combustion Products and Characteristics

2026-01-0293

4/7/2026

Accurate prediction of equilibrium combustion products and thermodynamic properties is essential for optimizing engine performance, enhancing combustion efficiency, and reducing emissions in diesel-powered systems. Traditional methods for combustion modeling often involve solving complex chemical equilibrium equations or thermodynamic relations, which could be computationally expensive and time-consuming. In this study, we present a data-driven approach using a deep neural network (DNN) model to predict the equilibrium combustion products and key thermodynamic characteristics of diesel under varying thermodynamic conditions. The proposed DNN model is trained on a comprehensive dataset generated from equilibrium calculations. The inputs include pressure, temperature, and equivalence ratio, covering a relatively wide range to encompass diesel equilibrium combustion under various conditions. Outputs are equilibrium combustion products and thermodynamic properties, including enthalpy

Ji, Huangchang, Wang, Kai, Guo, Zhefeng, Han, Yang, Lee, Timothy

Multi-Objective Reinforcement Learning Framework for Transmission Shift Schedule Optimization

2026-01-0162

4/7/2026

Building upon previous work that successfully employed a Reinforcement Learning (RL) agent for the autonomous optimization of transmission shift programs to enhance fuel efficiency, this paper addresses a critical limitation of that approach: the neglect of human-centric factors. While the prior methodology achieved substantial fuel consumption reductions by training an RL agent in a Software-in-the-Loop (SiL) environment, it did not explicitly account for aspects such as driver comfort and preferences, which are paramount for real-world user acceptance and drivability. This work presents a multi-objective optimization framework extending the artificial calibrator to simultaneously maximize fuel efficiency and enhance driver comfort. The method introduces a modified RL reward function that penalizes undesirable shift behavior to ensure a smooth driving experience (drivability). This new methodology also incorporates a mechanism to capture and integrate driver preferences, moving beyond

Kengne Dzegou, Thierry Junior, Schober, Florian, Rebesberger, Ron, Henze, Roman, Sturm, Axel

Machine Learning Based Meta-Analysis of IIHS Moderate Overlap Frontal Crash Test for Rear Seat Occupant Safety Performance

2026-01-0579

4/7/2026

A machine learning (ML)-based meta-analysis was conducted to evaluate rear seat occupant safety performance in the Insurance Institute for Highway Safety (IIHS) Moderate Overlap Frontal (MOF) 2.0 crash test. ML models were trained on historical IIHS crash test data to predict rear passenger injury metrics using vehicle architecture, restraint system characteristics, crash pulse parameters, and vehicle kinematics as input features. The models demonstrated high predictive accuracy and were subsequently used in a Sobol sensitivity analysis to identify critical design parameters influencing injury outcomes. The analysis revealed that rear passenger injury metrics were most sensitive to restraint system parameters. Specifically, crash pulse magnitude was the dominant factor for head injury metrics, pretensioner activation time for neck tension force, and lap belt force for the Neck Injury Criterion (Nij). For chest-related metrics—sternum deflection, dynamic belt position, and maximum belt

Lalwala, Mitesh, Kim, Wonhee, Furton, Lisa, Song, Jay

Machine Learning-Based Prediction of Tensile Properties in Cast Stainless Steel Influenced by Chemical Composition Variation

2026-01-0237

4/7/2026

In the category of cast stainless steels, there are several variants per different level of addition of chromium, vanadium along with some minor elements, such as molybdenum, niobium, tungsten to meet the requirement of corrosion and oxidation resistance. However, the influence of chemical composition variations on the mechanical properties of cast SS continues to lack a clear understanding. In the present study, via machine learning, the effects of each element on the tensile properties of the selected cast stainless steel are studied. The machine learning model is then used to predict how variations in elements affect tensile behavior, with the predictions validated through physical testing.

Mishra, Neelam, Biswas, Surjayan, V S, Rajamanickam, Aluru, Phani, Liu, Yi, Akbari, Meysam, Coryell, Jason

A Cooperative Hand Gesture Recognition System for Human Machine Interface to Control Robotic Arms

2026-01-0252

4/7/2026

In the context of Industry 5.0, effective human–machine collaboration requires seamless and natural interaction. Hand-Gesture Recognition (HGR) has emerged as a promising technology for developing human–machine interfaces (HMI) that enable users to control robotic systems without physical controllers or wearable devices. This research presents a real-time HGR system designed to control a 6-Degree-of-Freedom (DoF) robotic arm using YOLOv10, a state-of-the-art deep learning model for hand gesture detection and classification. While YOLOv10 delivers high recognition accuracy, its computational demands surpass the capabilities of edge devices typically mounted on robotic platforms, creating a hardware bottleneck. To address this challenge, a cooperative client–server architecture is proposed, distributing computational workload between the edge device and a more powerful remote server. An RGB camera attached to the robotic arm captures hand gesture images and transmits them to the server

DeHaven, Aaron Lee, Park, Jungme

From CFD Simulations to Machine Learning: A Comparative Study of ML Tools for Aerodynamic Drag Prediction

2026-01-0598

4/7/2026

This study presents a comparative assessment of two machine learning approaches for predicting aerodynamic drag coefficients (Cd) in automotive vehicle designs using data derived from computational fluid dynamics (CFD) simulations. The first approach employs traditional regression models trained on structured parametric data generated through controlled geometric variations, while the second approach integrates unstructured point-cloud geometry with structured metadata using a multi-modal deep learning framework. Both methods are evaluated within their respective contexts to understand their strengths, limitations and potential roles in automotive aerodynamic workflows. Rather than identifying a single best approach, the study highlights how these methods address different design needs and resource constraints, providing insights for future hybrid strategies that combine interpretability with geometric sensitivity. The work aims to establish a foundation for data-driven aerodynamic

Kumar, Gaurav, Khanna, Susheel

Leveraging Machine Learning for Occupant Body Size Estimation via B-Pillar Seatbelt Systems

2026-01-0559

4/7/2026

Occupant body size in vehicles varies significantly, encompassing differences in height, mass, and overall body composition. Adaptive restraint systems, featuring adjustable parameters such as belt load limiters, steering column load limiters and stroke, seat pan stiffness, and airbag pressure, can offer more equitable protection tailored to individual body sizes. In this study, a test rig modeled after the Volvo XC90 (2016) was used to collect data from 46 participants who were dressed in typical summer clothing and seated upright, without slouching or leaning sideways. Stepwise adjustments of the seat pan and seatback were performed. The collected measurements include seat pan movements (front-back and up-down), seatback recline, and key seatbelt-related parameters, such as belt payout length, D-ring angle, lap belt length, and buckle tension. The collected data was then used to train machine learning models to predict individual occupant characteristics: standing height, mass, and

Wang, Da, Ahmed, Jawwad, Rowe, Mike, Brase, Dan

A Real-Time Spatio-Temporal Configuration Framework for Shared Autonomous Electric Vehicle (SAEV) Fleets Integrating Deep Learning and Multi-Agent Simulation

2026-01-0035

4/7/2026

Shared Autonomous Electric Vehicles (SAEVs) can enhance urban mobility and efficiency. However, their operational performance is often hindered by the spatio-temporal imbalance between vehicle supply and passenger demand, leading to long wait times. This paper develops a novel repositioning framework where a lightweight CNN, informed by computationally intensive multi-agent simulations, enables real-time strategy deployment. The results show that: (1) An optimized repositioning policy, calibrated via multi-agent simulation, effectively cuts the mean passenger waiting time from 12.0 to 3.0 minutes (a 75% reduction). (2) A lightweight CNN surrogate model enables real-time deployment, reducing the policy computation time from ~4 hours to ~5 minutes (>98% faster). (3) The deep learning surrogate achieves this speed with a negligible performance trade-off, increasing the waiting time by only 0.156 minutes (4.9%) compared to the full optimization.

Shang, Kai, Wang, Ning

Machine Learning Modeling for Compressive Property Prediction of Foams

2026-01-0512

4/7/2026

Foam material models for automotive structural analysis typically require tensile and compressive data at multiple strain rates. The testing is costly and may require a long time to complete. For many applications, foams of similar chemistry are used and the foam structural responses, such as stiffness and compression force deflection, are controlled by the foam density. In such cases, Machine Learning (ML) lends itself as an ideal tool to detect the trends in material response based on density and strain rate. In this paper, two sets of polyurethane (PU) foams of different densities were tested at four strain rates ranging from 0.01/s to 100/s. ML models capable of predicting compressive stress-strain response for a range of densities were developed. The models demonstrated good prediction capability for intermediate strain rates at all foam densities and in extrapolating stress-strain curves at higher densities at all strain rates. The strain rate trends for density outside of the

M, Gokula Krishnan, Kavimani, Harish, Muppana, Sai Siddhartha, Savic, Vesna, Chavare, Sudeep, V S, Rajamanickam

CFD-Guided DoE-ML Optimization Methods in a Heavy-Duty Hydrogen Engine

2026-01-0326

4/7/2026

This study introduces a CFD-guided design of experiments (DoE) and machine learning (ML) framework for the co-optimization of piston and pre-chamber geometries in a passive pre-chamber heavy-duty hydrogen engine operating at medium and low loads. Starting from a reference configuration, an omega-type piston and a methane-optimized pre-chamber, the design space was parameterized using seven geometric variables. A Sobol sequence was employed to generate 96 randomized design variants in the DoE, each evaluated through high-fidelity 3D-CFD simulations to capture key combustion and performance metrics. The resulting dataset served as the foundation for developing and evaluating several ML regression models. A rigorous ML workflow was adopted, featuring 5-fold cross-validation and hyperparameter tuning via Bayesian optimization to ensure generalization and robustness. Model selection was based on multi-metric performance criteria including prediction accuracy, error stability, and

Menaca, Rafael, Shakeel, Mohammad Raghib, Liu, Xinlei, Mohan, Balaji, AlRamadan, Abdullah, Cenker, Emre, Silva, Mickael, Zhang, Anqi, Pei, Yuanjiang, Im, Hong

From Propane to Pi: Affordable AI-Driven Solutions for Sustainable Bus Heating

2026-01-0164

4/7/2026

This paper proposes an intelligent, artificial intelligence (AI) enabled seat heating system for school buses that saves energy by only activating heating elements when a passenger is identified. A custom-trained YOLOv8 deep learning model identifies passengers in real time and opens/closes real-time control of the individual electric seat heaters via a Raspberry Pi 5. The detector achieves around 10 frames-per-second (FPS) of inference on the Raspberry Pi 5 and 80–90 FPS on a laptop with over 92% detection confidence across various illumination conditions. Energy modeling shows the anticipated demand for a 10-kW propane-based heater is approximately 75% lower by implementing a 2.52 kW electric seat-heating system. In a typical operation schedule of 540 hours a year, this results in 4,000–5,000 kWh of annual savings, $465–$579 of annual cost savings and mitigates 0.9–1.3 t CO₂ per bus, annually. When implemented at the fleet level, the energy and cost saving will be in proportion. This

Chikkala, Daney Bhargav, Zadeh, Mehrdad, Tan, Teik-Khoon, Ponnam, Jitin, Batte, Jai Rathan

Ensemble of Experts for Generating Digital Parking Maps from Remote Sensing Images

2026-01-0107

4/7/2026

A digital parking map with precise parking spot geospatial information is crucial for tasks such as automatic valet parking, parking spot recommendations, and parking route optimization. This paper presents a parking map generation scheme that extracts high-definition parking spot geometry from remote sensing images. These images often suffer from occlusion, inconsistent resolution, and varying luminosity conditions. The proposed scheme utilizes a model ensemble paradigm, integrating multiple machine learning models to enhance the accuracy and quality of the generation of parking maps. The experiments demonstrate that the proposed scheme achieves an 80.5% parking spot detection precision and a center-to-center geometric representation error of 0.93 meters.

Shukla, Ajitesh, Cao, Xiaofei, Liu, Yongkang, Takeuchi, Yusuke, Sisbot, Akin

A Framework for Predicting Mechanical Performance of FFF Printed Parts Using Multimodal Deep Learning on Infrared Images

2026-01-0140

4/7/2026

Fused filament fabrication (FFF) has gained popularity in recent years because it can produce prototypes and functional components with complex geometry. Because of inherent process variability, the components often exhibit defects such as warping, layer delamination, voids, and poor surface finish, as well as issues related to variable material strength and anisotropy. In-situ monitoring (ISM) of the FFF process is a promising technique to predict part performance, which in turn can support accept or reject decisions for printed parts. This paper proposes a framework for incorporating ISM-generated information, with a particular focus on infrared (IR) image analysis for this purpose. IR camera images, in conjunction with numerical features such as infill pattern and extruder nozzle temperature, serve as an input to a multimodal deep learning (MDL) model that predicts the mechanical performance of printed parts. In the framework, convolutional neural nets process image inputs, while a

Mollan, Calahan, Kulkarni, Saurabh, Malik, Ali Ahmad, Patterson, Albert E., Pandey, Vijitashwa

Real-time Parameter Optimization for Eco-driving Control in Connected and Automated Vehicles Using Reinforcement Learning

2026-01-0041

4/7/2026

This paper introduces a novel methodology to enhance the energy efficiency of eco-driving controllers in Connected and Automated Vehicles (CAVs) by leveraging reinforcement learning (RL) techniques for real-time parameter optimization. Traditional eco-driving strategies rely on fixed control parameters, which limit adaptability across diverse traffic and road conditions. To address this, we apply continuous action space RL algorithms, specifically Deep Deterministic Policy Gradient (DDPG) and Proximal Policy Optimization (PPO), to dynamically tune four key parameters within a model predictive control framework that is grounded in Pontryagin’s Maximum Principle (PMP). These parameters influence acceleration, braking, cruising, and intersection-approach behaviors, making them critical for achieving optimal eco-driving performance. Our study employs Argonne National Laboratory’s RoadRunner simulator, a Simulink-based environment designed for high-fidelity CAV analysis, incorporating

Zhang, Yaozhong, Ammourah, Rami, Han, Jihun, Moawad, Ayman, Shen, Daliang, Karbowski, Dominik

Development of a Markov-Chain Prediction Model for a Work-Stealing Scheduler in Dynamic Scheduling for Multicore Embedded Software Systems

2026-01-0066

4/7/2026

Designing embedded software that achieves effective utilization of the fast-growing multicore embedded hardware should help to reduce their execution time and power consumption and improve their reliability. AI and machine learning algorithms are making their way into such rapidly enhanced multicore embedded hardware. We have developed a Markov-chain prediction model and integrated it into a work-stealing scheduler within a dynamic scheduling runtime layer (DSRL). Dynamic scheduling with a work-stealing scheduler was adapted from MIT’s Cilk framework [1]. Dynamic scheduling allows independent computations to be spawned so they can be scheduled dynamically and executed in parallel on available cores. Cilk used a random model in its work-stealing scheduler where an idle core randomly selects other cores to steal computations from them. However, Markov-chain-based scheduler allows idle cores to make informed decisions about which cores are better to steal their computation to increase

Sadeh, Waseem, Ganesan, Subramaniam, Qu, Guangzhi, Rawashdeh, Osamah

AI-Enhanced Functional Safety in ADAS Controllers: Predictive Fault Management under ISO 26262

2026-01-0036

4/7/2026

Ensuring ISO 26262 functional safety in advanced driver assistance systems (ADAS) is increasingly complex as these platforms integrate artificial intelligence (AI) for perception, decision-making, and vehicle control. Traditional safety mechanisms are largely deterministic, but AI introduces non-determinism, creating challenges for verification, validation, and certification. Real-time vehicle telemetry, sensor outputs, and environmental inputs are processed through machine learning algorithms that forecast hardware and software faults before they escalate into hazardous conditions. These predictions are systematically integrated with ISO 26262 safety measures, enabling adaptive diagnostics, fault isolation, and rapid recovery strategies. The AI model introduces hazards such as data bias, model drift, opaque decision-making, and unsafe automation. A dedicated AI Hazard Analysis and Risk Assessment addresses data quality, validation, monitoring, explainability, and fail-safe mechanisms

Abdul Karim, Abdul Salam

Developing Safety Case for Automated Driving following UL 4600 & ISO 5083

2026-01-0527

4/7/2026

Automated Driving Systems (ADS) rely on AI algorithms, machine learning, and sensor fusion to perform autonomous driving tasks. Safety challenges arise due to the probabilistic behavior of AI/ML algorithms and the need to ensure safety within defined Operational Design Domains (ODDs). Traditional standards such as ISO 26262[3] (Functional Safety) and ISO 21448[4] (SOTIF) address hardware and software failures or functional deficiencies but are insufficient for higher-level autonomous systems (SAE Levels 3–5). To close this gap, additional standards such as UL 4600[1] and ISO 5083[2] provide complementary frameworks for ADS safety assurance. UL 4600[1] establishes a claim-based safety case encompassing the vehicle, infrastructure, and processes, emphasizing structured arguments supported by evidence and reasoning. It offers guidance on autonomy functions, V & V, tool qualification, dependability, and safety culture. ISO 5083[2] focuses on design, verification, and validation of ADS

Mudunuri, Venkateswara Raju, Almasri, Hossam, Fan, Hsing-Hua

Roundabout Dilemma Zone Data Mining and Forecasting with Trajectory Prediction and Graph Neural Networks

2026-01-0029

4/7/2026

Traffic roundabouts, as complex and safety-critical road scenarios, present significant challenges for autonomous vehicles. In particular, predicting and managing dilemma zone (DZ) encounters at roundabout intersections remains a pivotal concern. This paper introduces an AI-driven system that leverages advanced trajectory forecasting to anticipate DZ events, specifically within traffic roundabouts. At the core of our framework is a modular, graph-structured recurrent architecture powered by graph neural networks (GNNs). By modeling agent interactions as a dynamic graph, our approach integrates heterogeneous data sources - including semantic maps - while capturing agent dynamics with high fidelity. This GNN-based forecasting model enables accurate prediction of DZ events and supports safer, data-driven traffic management decisions for both autonomous and human-driven vehicles. We validate our system on a real-world dataset of roundabout intersections, where it achieves high precision

Lu, Duo, Satish, Manthan, Farhadi, Mohammad, Chakravarthi, Bharatesh, Yang, Yezhou

Assessing Plug-in Electric Vehicle Adoption: Methodologies, Policy Effects, and Diverging Market Pathways in China, the U.S., and Europe

2026-01-0455

4/7/2026

Accurate projection of Plug-in Electric Vehicle (PEV) market sales share is vital for evidence-based policymaking, yet existing studies employ diverse and often fragmented methodologies, creating a need for a systematic review to clarify their analytical foundations and comparative strengths. This study classifies mainstream approaches to market projections into theory-driven and data-driven categories and reviews the merits, limitations, and future directions of five representative models. Analysis reveals that leading approaches increasingly employ cross-scale model coupling, theory-data fusion, and modular design to harness complementary strengths, improving model robustness and predictive accuracy. Furthermore, the study compares PEV policies and market outlooks in China, the United States, and Europe—the world's three largest automotive markets. The findings indicate a strong linkage between projection convergence and policy stability. China demonstrates the highest policy

Luo, Wei, Ou, Shiqi(Shawn), Zhou, Pan, Wang, Tianpeng, Qian, Xiaodong

CAE Acceleration with Machine Learning for Results Prediction

2026-01-0488

4/7/2026

Industries are following a tedious product development cycle for developing their product. In product development major steps includes design ideas, Drawings, CAD, CAE, Testing and design improvement cycle. This is a monotonous process and takes time which impacts on its time to deliver product and cost on development. Now a days industries are fast growing and targeting to reduce development cycle time and cost. AI&ML is impacting almost all areas in the industry and significantly reducing efforts time and cost. To make use of AI&ML in CAE, Altair Physics AI is an effective tool. To ensure the design of product traditional way is to develop a CAD of the product, develop, perform CAE and analyze performance. If we consider CAE procedure it is time consuming process which includes FEA model build, applying boundary conditions, running simulation and analyzing results which could take minutes to hours. By using ML with Physics AI we can make predictions on new design of the product in

Dangare, Anand Manohar, Kulkarni, Mandar

Neural Surrogates and Reinforcement Learning for Heavy-Duty Diesel Torque Control

2026-01-0199

4/7/2026

Accurate torque-trace reproduction on regulatory drive cycles is central to heavy-duty diesel certification and development testing. Conventional controllers such as Proportional Integral Derivative (PID or PI) can be enhanced with gain scheduling and feedforward (FF) maps to satisfy requirements but require extensive calibration and are sensitive to nonlinearities and delay. This paper evaluates a data-driven control framework comprising a recurrent neural surrogate of engine torque (specifically an LSTM – long short-term memory) trained on engine/dynamometer data and a reinforcement learning (RL) policy trained using this surrogate (“world model”) to track requested torque while regularizing control effort. The RL policy (specifically TD3 – twin delayed deep deterministic) is benchmarked against tuned PID and PID+FF baselines on the Environmental Protection Agency’s Heavy Duty Federal Test Procedure (HD-FTP) segments using EPA regression criteria (slope, |intercept|, R2) and tracking

Cook, James, Puzinauskas, Paulius, Bittle, Joshua, Hall, Spencer

A Frequency-Aware Reinforcement Learning Framework for Suspension Control with Integrated Temporal Features and Expert Guidance

2026-01-0209

4/7/2026

Active suspension systems play a crucial role in improving vehicle ride comfort and handling stability. However, most existing studies focus on the low-frequency range below 20 Hz, leaving the suppression of high-frequency vibrations within 50–500 Hz largely unexplored, even though these vibrations strongly affect in-cabin noise and ride quality. To address this gap, this study introduces a quarter-car suspension model incorporating both bushing dynamics and a rigid-ring tire within a reinforcement learning (RL) framework. A major challenge for RL-based suspension control is its degradation in high-frequency performance. To overcome this issue, we design an innovative training framework that integrates multiple synergistic strategies. First, frequency-domain rewards are incorporated as auxiliary signals to explicitly guide policy optimization in the high-frequency band. Second, long short-term memory (LSTM) networks are embedded in both the Actor and Critic to capture the sequential

zhu, Zhehui, Zhang, Lijun, Meng, Dejian, Hu, Xingyu

Integrated DBSCAN-Based Segmentation of Tractor Activity into Productive and Non-Productive States from GPS Data

2026-01-0169

4/7/2026

Accurate identification of Productive and Non-Productive States or tractor duty cycles—comprising working, idle, and transport states—is critical for performance analysis, fuel optimization, and emissions modeling in agriculture machinery and fleet monitoring. This study explores the application of integrated unsupervised machine learning (ML) techniques to classify duty cycles using GPS-derived parameters such as speed, location variance, and temporal patterns. Unlike supervised approaches, the proposed method does not rely on several labeled engine and vehicle parameters, making it scalable and adaptable across diverse operational contexts. Clustering algorithm DBSCAN (Density-Based Spatial Clustering of Applications with Noise) in integration with hybrid rule-based and a road feature is employed to segment GPS data into distinct behavioral states. Feature engineering focuses on extracting motion signatures and spatial-temporal features that correlate with operational modes

Maharana, Devi prasad, Gangsar, Purushottam, Dharmadhikari, Nitin, Pandey, Anand Kumar

Modular Smart Corner Systems for Next-Generation Electric Vehicle Architecture

2026-01-0639

4/7/2026

The transition to software-defined vehicles (SDVs) necessitates a paradigm shift in both control strategies and vehicle architecture. The EU-funded R&D project SmartCorners addresses this challenge by developing integrated, modular, and scalable smart corner systems (SCS) that combine in-wheel motor (IWM)-based propulsion, brake blending, active suspension system, and steer-by-wire functionality in one module. These SCS can be retrofit or smoothly integrated into the highly adaptable skateboard chassis architecture of modern electric vehicles (EVs), enabling scalable deployment across diverse vehicle types. The central approach of this paper is the utilization of artificial intelligence (AI) and machine learning (ML) to implement multi-layer, data-driven control strategies, facilitating real-time actuation, fault mitigation, and user-centric EV architecture. The SmartCorners project strives to demonstrate significant enhancements, including improved real-world driving range due to

Ratz, Florian, Armengaud, Eric, Formento, Cecilia, Moscone, Giulia, Sorrentino, Gennaro, Bisciaio, Giorgio, Sorniotti, Aldo, Amati, Nicola, Braun, Daniel, Deibler, Bernd, Boxberger, Valerius, Sottile, Salvatore, Ivanov, Valentin, Fuse, Hiroyuki, Kompara, Tomaž

A Dataset for Visual Classification of Bulging Pouch Battery Cells

2026-01-0393

4/7/2026

The rapid advancement of lithium-ion battery technologies, particularly pouch cells, has driven significant growth in electric vehicles, mobile devices, and renewable energy storage. However, pouch cells are especially susceptible to mechanical deformation and failure, including bulging caused by internal gas formation—a common indicator of cell aging or imminent failure. In this study, we developed a visual dataset of bulging pouch battery cells to support real-time diagnostics and safety monitoring in industrial and laboratory environments. The dataset includes 200 high-resolution images (100 bulged, 100 normal) curated through a web-crawling and filtering pipeline. The dataset is benchmarked across several traditional machine learning models to evaluate performance and feasibility for edge AI deployment. The best model achieved strong classification accuracy while maintaining a small computational footprint suitable for embedded applications.

Alkawasmie, Mohammad, Farooqui, Saad, Algalham, Dheya, Rahman, Mahfilur, Challa, Karthikeya, Maxim, Bruce, Shen, Jie

Virtual Sensor Development for Real-Time Estimation of Transient NOx Emissions for Internal Combustion Engine

2026-01-0370

4/7/2026

Accurately measuring NOx emissions under transient engine conditions is becoming increasingly important with upcoming Euro 7 and EPA 2027 regulations. Traditional physical sensors often struggle with cost and response time, especially with aging of sensors in dynamic operation. This paper introduces a machine-learning–based virtual NOx sensor that can provide real-time emission estimates while reducing reliance on hardware sensors. The approach uses multiple machine-learning methods (Random Forest, Bootstrap Aggregating, Adaptive Boosting, Gradient Boosting, Extreme Gradient Boosting) and selected best one to establish correlations between engine operating parameters, measured steady-state data, and transient duty cycle NOx emissions. Validation across different duty cycles has shown strong alignment with physical sensor readings, with R2 values above 99.95% for training cycle data sets and above 95.34% for held-out cycles during training. The model needs to be trained with larger

Kumar, Chandan, Dahodwala, Mufaddel, Thawrani, Kiran

Curvature-Based End-to-End Autonomous Vehicle Path Planning at Intersections

2026-01-0045

4/7/2026

Autonomous vehicle navigation requires accurate prediction of driving path curvature to ensure smooth and safe trajectory planning. This paper presents a novel approach to curvature prediction using deep neural networks trained on GPS-derived ground truth data, rather than model predictions, providing a more accurate training signal that reflects actual vehicle motion. We develop a multi-modal neural network architecture with temporal GRU encoders that processes vision features, driver intent signals, historical curvature, and vehicle state parameters to predict curvature. A key innovation is the use of GPS-based actual curvature measurements computed from vehicle motion data (κ = ωz/v) as training supervision, enabling the model to learn from real-world driving patterns. The model is trained on 5,322 samples from real-world driving data collected on The University of Oklahoma’s Norman Campus using a Comma 3X device and a 2025 Nissan Leaf electric vehicle. Experimental results

Hajnorouzali, Yasaman, Wang, Hanchen, Li, Taozhe, Burch, Collin, Lee, Victoria, Tan, Lin, Arjmandzadeh, Ziba, Xu, Bin

In-situ Decoupling and Stiffness Optimization of Resiliently Coupled Assemblies Using FRF-Based Substructuring

2026-01-0227

4/7/2026

Resilient mounts are critical in controlling vibration transfer from sources such as engines, motors, and suspension to the vehicle structure. Conventional optimization methods rely on finite element analysis (FEA), which, while accurate, is computationally intensive and limits iterative NVH development. This paper introduces a Frequency Response Function Substructuring (FBS)-based approach that decomposes the system into substructures characterized by FRFs, significantly reducing computational cost without compromising accuracy. Key contributions include: (1) recovering subsystem FRFs from coupled system data in-situ for mount optimization, (2) extending FBS to handle enforced motion, and (3) proposing an alternative strategy for cases with unknown or unmeasurable loads. The methodology is demonstrated on a mid-size pickup truck model to optimize seat track response under a Four post shake load by refining body mounts. These advances broaden the applicability of FBS for efficient NVH

Haider, Syed, Abbas, Ahmad, Jahangir, Yawar, Maddali, Ramakanth

Adaptive Economic Model Predictive Control for Engine Emissions Management with Compensation for Performance Drift

2026-01-0287

4/7/2026

This paper addresses the changes in engine emissions due to in-use component changes through the synergistic application of predictive control, machine learning, and onboard adaptation. In particular, we consider an adaptive economic Model Predictive Control (eMPC) strategy to mitigate the effects of performance drift on Nitrogen Oxides (NOx) and Soot emissions from compression ignition (diesel) engines. A performance drift block, which applies a multiplier and offset to nominal emissions, is integrated with a high-fidelity Neural Network (NN) plant model to simulate these characteristic changes. To counteract variability, two online adaptation methods are integrated within the eMPC framework: One is based on Recursive Least Squares (RLS) and another on a continuously updated online NN. The proposed control architecture is validated through simulations over standard transient cycles. Results demonstrate that while the rate-based eMPC possesses inherent robustness to performance drift

Zhang, Jiadi, Li, Xiao, Kolmanovsky, Ilya, Tsutsumi, Munecika, Nakada, Hayato

Study on the Regulation of the Three-Phase Microenvironment in Low-Platinum MEA

2026-01-0443

4/7/2026

With the growth of energy demand, fuel cells as efficient and clean energy devices, have attracted increasing attention. However, the high cost of membrane electrode assembly (MEA) restricts their large-scale application. Therefore, reducing the platinum usage and improving performance have become key research point. In this work, MEA was prepared and excellent performance of 1.52 W·cm-2 was achieved at a low platinum loading. The influence of different ionomer/carbon (I/C) ratio on the performance of fuel cells was systematically investigated. It was found that the performance of the MEA was the highest when the I/C ratio is 0.6. Quantifying hydrophilic and hydrophobic characteristics of catalyst layers with varying ionomer contents revealed that the proton conduction efficiency is optimal when the I/C ratio is 0.6. This balance established efficient proton conduction pathways, from the results of proton conduction impedance testing. SEM analysis demonstrated that pore structure

Li, Xin, Cai, Xin, Lin, Rui

Items per page:

1 – 50 of 1443