Browse Topic: Neural networks

Items (532)
Find
Optimisation of frontal impact force scheme to accelerate the target setting process for development of front-end structures in vehicles.2026-26-00011/16/2026
A passenger vehicle's front-end structure's structural integrity and crashworthiness are crucial to ensure compliance with various frontal impact safety standards (such as those set by Euro NCAP & IIHS). For a new front-end architecture, design targets must be defined at a component level for crush cans, longitudinal, bumper beam, subframe, suspension tower and backup structure. The traditional process of defining these targets involves multiple sensitivity studies in CAE. This paper explores the implementation of Physics-Informed Neural Networks (PINNs) in component-level target setting. PINNs integrate the governing equations into neural network training, enabling data-driven models to adhere to fundamental mechanical principles. The underlying physics in our model is based upon a force scheme of a full frontal impact. A force scheme is a one-dimensional representation of the front-end structure components that simplifies a crash event's complex physics. It uses the dimensional and
Gupta, Ishan, Bhatnagar, Abhinav, Kumar, Ayush
A Mathematical Modelling Approach Based on Artificial Neural Network for Estimation of Key Parameters in Internal Combustion Engine2026-26-06591/16/2026
In a conventional powertrain driven by Internal combustion (IC) engines, there are many key sensors such as intake manifold temperature sensor, exhaust temperature sensor, Nitrogen oxide (NOx) sensors etc. to monitor engine performance and emissions. For enabling various diagnostics strategies and engine monitoring, virtual sensors or modelled values of such sensors play an important role. Conventional strategies incorporate the use of regression models, map-based models and physics-based models. There are a few drawbacks with conventional models in terms of accuracy and model calibrations efforts. Data driven models or neural networks have fairly better accuracy and reliability for estimating complex parameters. Representing the neural network with a mathematics-based model would help to eliminate drawbacks associated with conventional modelling approach. The proposed methodology uses artificial intelligence technique called artificial neural network (ANN) for estimation, for example
Jagtap, Virendra Shashikant, Shejwal, Sanket, Mitra, Partha
Driving behavior modelling with Graph Attention Networks using spatial data2026-26-06611/16/2026
In automotive engineering, understanding driving behavior is crucial for decision on specifications of future system designs. This study introduces an innovative approach to modeling driving behavior using Graph Attention Networks (GATs). By leveraging spatial relationships encoded in H3 indices, we construct a graph-based model that captures dependencies between various vehicle operational parameters and their operational regions using H3 indices. The model utilizes CAN signal features such as speed, fuel efficiency, engine temperature, and categorical identifiers of vehicle type and sub-type. Additionally, regional indices are incorporated to enrich the contextual information. The GAT model processes these heterogeneous features, learning to identify patterns indicative of driving behavior. This approach offers several significant advantages. Firstly, it enhances the accuracy of driving behavior modeling by effectively capturing the complex spatial and operational dependencies
Salunke, Omkar
A Physics-Guided Neural Network Approach for EGR Cooler Temperature Estimation: Addressing ECU Model Discrepancies2026-26-04351/16/2026
As the automotive industry explores alternative powertrain options to curb emissions, it is pertinent to refine existing technologies to improve efficiency. The Exhaust Gas Recirculation (EGR) system is a pivotal component in emission control strategies for Internal Combustion Engines (ICE). The EGR cooler is crucial in thermal management strategies, as it lowers the temperature of recirculated exhaust gases before feeding it along with fresh air, thereby reducing nitrogen oxides (NOx) emissions. Precise estimation of the EGR cooler outlet temperature is crucial for effective emission control. However, conventional Engine Control Unit (ECU) models fall short, as they often show discrepancies when compared to real-world test data. These models rely on empirical relationships that struggle to capture, transient effect, and real time variation in operating conditions. Consequently, ECU-based estimations of EGR Cooler outlet temperature frequently deviate from actual values, leading to
Kumar, Amit, Kumar, Raman, Manojdharan, Arjungopal, Challa, Krishna, Kramer, Markus
Extreme Learning Machines (ELM) for Fast Instantaneous Prediction of Emission Level of an Automobile2026-26-02261/16/2026
Addressing climate issues is a key aspect of good global governance today. A key aspect of managing the threats caused to the environment around is to ensure a sustainable transportation system so that humans exist in peace with nature. According to sources, in 2020 alone, cars accounted for approximately 23% of global CO2 emissions. In addition, they also emit nitrogen oxides and other pollutants thus damaging the ecosystem. In line with this objective, there are emission level testing strategies in place in each country. However, we can do better for a sustainable future. On one hand, the huge volume of vehicles around the world makes it an excellent choice and source for a vast emission level dataset comprising of features such as vehicle type, fuel type, driving conditions, road type etc.. as well as the target variable representing the emission band of the vehicle. In addition to the big data available as mentioned above, there is a marked progress in the machine learning
Sridhar, Sriram, Aswani, Shelendra
Artificial Intelligence in Passenger Cars: Unlocking Intelligent Mobility and Human-Centric Vehicle Innovation2026-26-06721/16/2026
Artificial Intelligence (AI) is fundamentally reshaping the landscape of automotive engineering, particularly in passenger vehicles where comfort, safety, and customization are paramount. This paper presents a comprehensive review of current and emerging AI-driven applications in passenger cars, including intelligent vehicle insurance, real-time traffic optimization, personalized driving experiences, emotion recognition, and advanced EV battery management. The integration of AI technologies—ranging from machine learning and computer vision to natural language processing and deep neural networks—enables predictive maintenance, adaptive driver-assist systems, and next-generation human-machine interfaces. AI’s role in vehicle insurance is also explored through behaviour-based risk scoring and claims automation. Furthermore, emotion-aware cabin systems are discussed, which aim to detect driver fatigue, stress, or distraction to improve road safety and mental wellness. The paper also
Hazra, Sandip, Tangadpalliwar, Sonali, Khan, Arkadip
Transforming Gear Manufacturing_ A Deep Learning Approach to Quality Prediction2026-26-06461/16/2026
The automotive industry is rapidly transitioning towards Industry 4.0, transforming vehicle manufacturing. To achieve a lower carbon footprint, it is crucial to minimize raw material wastage and energy consumption. Reducing component wastage, lead time, and automating gear manufacturing are key areas. Gear micro-geometry inspection is vital, as variations affect service life and NVH (Noise, Vibration, Harshness). Despite standards for permissible errors, manual evaluation is often needed.This subjective evaluation approach will have a possibility that a gear with undesired variations gets assembled into the product. These issues can be detected during NVH testing, leading to replacement of part and re-assembly thus increasing lead time. This generates a need for an automated system which could reduce the human intervention and perform the activity. This paper discusses methodologies to meet these requirements and achieve desired results. The research aims to develop a deep learning
Ramakrishnan, Gowtham Raj, Baheti, Palash, PR, Vaidyanathan, Durgude, Ranjit, Bathla, Archana, R, Greeshmita, V, Rangarajan
Regenerative braking control strategy for electric commercial bus under Indian road condition using artificial neural network2026-26-06451/16/2026
The main objective of this research is to create a more efficient regenerative braking control strategy for electric commercial buses operating under Indian road conditions. The strategy uses Artificial Neural Networks (ANNs) to optimize the regenerative braking process. Regenerative braking helps to recover energy that would otherwise be lost during braking and convert it back into usable power for the vehicle. The challenge is to design a system that works effectively on the diverse and often challenging road conditions found in India, such as varying gradients, traffic patterns, and road surface types. This study begins by collecting data (which includes vehicle speed, road gradient, traffic condition, etc.) from real-world driving conditions and aims to train an Artificial Neural Network (ANN) using a large set of driving data collected under various conditions to predict the most efficient regenerative braking settings for different driving scenarios. While the results are
Saurabh, Saurabh, Bhardwaj, Rohit, Patil, Nikhil, Gadve, Dhananjay, Amancharla, Naga Chaithanya
Artificial Neural Network for predicting thermal cool-down behavior of a School Bus Cabin2026-26-06531/16/2026
Passenger vehicles, such as buses, exhibit significant heat absorption when parked under direct sunlight, leading to elevated interior temperatures during daytime and a consequent reduction in thermal comfort. In our previous study, we developed and validated a transient numerical method to predict the thermal behavior of the cool-down process in a school bus cabin. Despite its accuracy, modeling such complex simulations using Computational Fluid Dynamics (CFD) remains time-intensive, with pre-processing taking up to a week and simulations requiring an additional 8-10 hours. While physics-based simulations are known for their accuracy, they are inherently slower. Conversely, data-based meta models offer faster results, but their outputs must be carefully correlated with test data or physics-based model data. This study aims to leverage existing CFD data to train and develop an Artificial Neural Network (ANN) based meta model, which is then benchmarked against test results for
Suryavanshi, Apoorva, sharma, shantanu
Remaining Charging Time Accuracy Improvement through Preconditioning Prediction in Electric Vehicles2026-26-06781/16/2026
Remaining Charging Time (RCT) is an essential metric that is displayed to the customer/driver while charging the vehicle. It tells the driver how much time is remaining for the vehicle to be charged to 100% Display State of Charge (SOC) or the target SOC set by the driver. This metric helps drivers plan their time and decide when to unplug their device. For large battery systems like electric vehicles, the understanding of how much longer the battery needs to charge can help optimize energy use and avoid overcharging. The avoidance of un-necessary prolonged charging affects the long-term health of the battery. The RCT is calculated using a basic formula as the ratio of total energy needed to charge the battery to target SOC and the power pushed into the battery for charging. In this RCT formula numerator for a battery to be charged to a specific target SOC is fixed, but the denominator depends on lot of factors during charging and Battery Preconditioning is one of those factors that
Jawed, Shahid, Srikantaiah, Sudarshan
NN FMU: Deep Learning Models for Next-Gen xiL Vehicle Simulation2026-26-04521/16/2026
Functional Mock-up Units (FMUs) have become a standard for enabling co-simulation and model exchange in vehicle system development. However, traditional FMUs derived from physics-based models can be computationally intensive, especially in scenarios requiring real-time performance. This paper presents a Python-based approach for developing a Neural Network (NN) based FMU using deep learning techniques, aimed at accelerating vehicle simulation while ensuring high fidelity. The neural network was trained on vehicle simulation data and developed using Python frameworks such as TensorFlow. The trained model was then exported into FMU, enabling seamless integration with FMI-compliant platforms. The NN FMU replicates thermal behavior of vehicle with high accuracy while offering a significant reduction in computational load. Benchmark comparisons with Physical Thermal Model demonstrate that the proposed solution provides both efficiency and reliability across various driving conditions
Srinivasan, Rangarajan, ASHOK BHARDE, Pooja, MHETRAS, Mayur, CHEHIRE, Marc
Accelerating engine auxiliaries’ development by deploying ML techniques in CFD virtual verification2026-26-03741/16/2026
The automotive and off-road industries are heavily investing in research and development (R&D) to improve both physical and virtual verification and validation techniques. Recent software and hardware advancements have extended these techniques from simple component evaluations to complex system assessments such as involving multi-physics scenarios. Despite the benefits of virtual validation tools like structural analysis and computational fluid dynamics (CFD), they often come with high development costs, particularly in CFD applications. Virtual verification methodology, especially when combined with data science, offer significant advantages over traditional physical methods by enhancing computer-aided engineering (CAE) efficiency and reducing resource consumption which can greatly improve product design and validation efficiency across many industries. The success of machine learning applications depends on effective data processing, adequate computational resources, and the right
Jadhav, Mitali, Kumbhar, Appaso, Tirumala, Bhaskar, Nisha, Kumari
Multi-Model Analysis of O-Ring Compression and Relaxation Force Predictions Using Machine Learning and CAE Simulations2026-26-04661/16/2026
O-rings play a critical role in ensuring leak-proof seals in a wide range of engineering systems. Accurate prediction of their compression and relaxation behavior under various material and geometric configurations is essential for optimal design and reliability. This study presents an analysis of machine learning techniques to predict two key performance outputs, compression force and relaxation force (after 10 minutes) trained on computer-aided engineering (CAE) simulation data. The experimental setup was represented in CAE simulation and the results were compared with experimental data conducted at ZF test facilities. Simulation results correlated well with the experimental data (deviation was less than the 5%). To create a dataset for training machine learning (ML) models, realistic ranges for the input parameters such as hardness and geometrical parameters were determined, and simulation data were generated using design of experiments (DOE). Multiple ML models were developed and
Kosgi, Durgaprasad, Alva, P Pancham, Dangeti, VenkataKrishna Pavan
A Deep Learning Approach for Dynamic Range Estimation in Electric Fleet Vehicles.2026-26-06751/16/2026
Keywords: Real-Time Range Prediction, Long Short-Term Memory (LSTM), Time Series Modelling, Deep Learning, Vehicular Telemetry, CAN Data, Range Anxiety Accurate and real-time range estimation is a critical enabler for efficient energy management and operational planning in Commercial Electric Vehicles (CVEVs). This project presents a robust deep learning-driven framework for vehicle range prediction, leveraging a Long Short-Term Memory (LSTM) network trained on high-frequency powertrain CAN data obtained from CVEVs operating across diverse driving environments. The dataset comprises key dynamic and environmental parameters—including State of Charge (SOC), vehicle speed, acceleration, and temperature—captured during real-world operations. Extensive data pre-processing techniques were applied to ensure signal integrity, including high-resolution timestamp synchronization, noise filtering, and imputation of missing values, thereby establishing a reliable foundation for time-series
surlekar, Samarth, Ghorpade, Swapnali, Kalghatgi, Bhakti
Implementation of new educational technologies for training personnel for the transport industry2026-26-02411/16/2026
Today's challenges affect the system of training highly qualified personnel for the transport industry, which is actively transforming through the introduction of innovative educational technologies. Key areas have become digitalization of the educational process, integration of virtual simulators, augmented reality technologies (AR/VR) and platform solutions for managing educational programs. Today, transport education in Russia is focused on advanced training of specialists capable of working with autonomous transport and neural network solutions. Development prospects are associated with further integration of adaptive educational trajectories and global network platforms to ensure the competitiveness of graduates in the digital economy. Taking into account the specifics of the modern Russian transport industry, using an analytical method, the article examines solutions to some priority problems, namely: • Methods of modernizing educational programs through designing «professions of
SHISHANOV, Sergei, Kurmaev, Rinat, revenok, Svetlana
Artificial Intelligence in Aeronautical Systems: TaxonomyAIR6987 (Current)11/26/2025
Establish a comprehensive taxonomy of Artificial Intelligence in aviation
G-34, Artificial Intelligence in Aviation
testing revision status on details pageCUR1513-jms36/3/2025
cur1580
Heavy-Duty Electric Vehicle Charging Consortium, Automated Vehicle Safety Consortium, International Alliance for Mobility Testing Standardization
revision 2jms_bp_24/8/2025
revision 2
Automated Vehicle Safety Consortium
Analysis of Left-Turn Mixed Traffic Flow Dynamics at Y-Shaped Signalized Intersections Using Cellular Automata2025-01-72042/21/2025
Through the method of on-site video observation, this study divides the intersection area into three parts according to the road traffic characteristics of the Y-shaped signalized intersections, and at the same time obtains the relevant parameters. These parameters include the left-turn speed and traffic density of motor vehicles within both the internal and exit areas, the frequency of lane-changing and queuing behaviors of non-motorized vehicles in the internal area, and the left-turn speed and traffic density of non-motorized vehicles in both the internal and exit areas. The data extraction and analysis of the parameters provide strong data support for further analysis of the subsequent mixed traffic flow. A cellular automaton model is developed using the intersection’s exit area as the scenario. The exit area is divided into three lanes based on the queuing patterns of mixed traffic. Corresponding traffic rules are established according to the traffic density of motorized and non
Yuan, Li, Liu, Xiaowei
Research and Simulation Implementation of Self-Driving Vehicle Trajectory Optimization in Networked Intersection Environment2025-01-71942/21/2025
Nowadays, the rapidly developing Connected and Autonomous Vehicle (CAV) provides a new mode of intersection vehicle cooperative control, which can optimize vehicle trajectories and signal phases in real time and reduce intersection delays through the advantages of vehicle-road cooperative information interaction and the high controllability of CAV. In this paper, the intersection of Jintong West Road and Guanghua Road in Beijing is taken as the research object, and the vehicle trajectory constraints, acceleration constraints, speed constraints, safe driving constraints, signal switch constraints and traffic signal control constraints are set up with the minimization of traffic delay as the objective function. The DQN deep reinforcement learning network is constructed based on vehicle states, vehicle actions, reward functions, and update rules, and starts learning and updating to generate the target network. Then, SUMO software is used to simulate and test and compare the trajectory
Xu, Yuting, Zhang, Yong, Wu, Xianyu
Biohybrid Robotic Hand Research PlatformTBMG-5215912/1/2024
Brain-machine interfaces enable direct communication between a brain’s electrical activity and an external device such as a computer or a robotic limb that allows people to control machines using their thoughts. Researchers have developed a novel biohybrid neuroprosthetic research platform comprised of a dexterous artificial hand electrically interfaced with biological neural networks.
Enabling Robots to Interpret Surroundings Quickly and AccuratelyTBMG-5218212/1/2024
Imagine having to straighten up a messy kitchen, starting with a counter littered with sauce packets. If your goal is to wipe the counter clean, you might sweep up the packets as a group. If, however, you wanted to first pick out the mustard packets before throwing the rest away, you would sort more discriminately, by sauce type. MIT engineers have developed a method that enables robots to make similarly intuitive, task-relevant decisions.
Enhancing Cargo Transportation using Intelligent systems for better Logistics Management2024-26-01831/16/2024
Efficient cargo transportation plays a crucial role in logistics management and supply chain operations. Accurately detecting and utilizing cargo space within vehicles is vital for maximizing transport capacity, minimizing costs, and optimizing resource allocation and time management. This research paper focuses on enhancing cargo utilization using intelligent systems to improve logistics management. The major research is on developing a system that combines computer vision algorithms and intelligent systems to detect and implement a combination of features for efficient use of cargo space within vehicles and monitor cargo to reduce losses. The proposed approach will use and utilize image processing methods to get relevant features and identify cargo areas. Machine learning models, such as convolutional neural networks (CNNs) and object detection frameworks, will be trained and evaluated on a comprehensive dataset of cargo images to identify the most effective approach for cargo space
JAGTAP, Pramod Prakash, Kale, Shreeja, Mahali, Rakesh
ANN Based Surrogate Model for Fillet Stress Prediction of Automotive Crankshafts2024-26-02821/16/2024
Designing of automotive crankshaft is a time-consuming process and often involves several iterations of crankshaft designs being validated by calculations till a safe design is conceptualized. In the initial phases of design process, Finite Element Analysis (FEA) methods are conventionally used for design validation of crankshaft by stress and fatigue calculations. Setting up of FEA models to find crankshaft fillet stress is laborious and often the design is required to undergo a few iterations of changes before it can be deemed safe for the next phases of development. This study aims to implement an Artificial Neural Network (ANN) prediction model to predict crankshaft fillet stress under a 3-point bending scenario and evaluates its performance. The objective of this study was to develop a Machine learning prediction model based on Artificial Neural Networks that predicts maximum von mises stress at Crankshaft fillets region. A dataset containing previous FEA stress data was used to
Thokalath, Jaison Jacob
Occupant Injury Severity Prediction in Road Traffic Accidents using Machine Learning Techniques2024-26-00111/16/2024
Over years, automotive industry has made a significant progress in improving passenger car safety systems to reduce occupant injuries and fatalities due to road accidents. To further enhance the vehicle safety, it is very important to have a deeper understanding on real world accidents and true safety benefits associated with various safety systems. This requires a framework to understand the effectiveness of safety systems towards reduction in occupant injury and fatalities. This study focusses on utilizing machine learning techniques to predict occupant injury severity by considering accident parameters and safety systems, using the Road Accident Sampling System - India (RASSI) real world accident data. The RASSI database contains comprehensive accident data, including various factors that contributes for occupant injury. The study focused on fifteen accident parameters that represent key aspects of crash scenarios such as vehicle type, accident type, vehicle speed, and occupant
G, Santhosh Kumar, Khatavkar, Akshay, Sahu, Dilip, Koralla, Sivaprasad, Kulkarni, Prasad
Helmet detection in 2-Wheeler2024-26-00071/16/2024
Wearing Helmet is a critical safety measure not only for riders but also for passengers. However, people often tend to skip wearing these protective headgears, thereby leading to, an increased risk of injury or death in the event of an accident. There is a growing necessity to develop innovative methods that automatically monitor and prevent unsafe driving. To address this issue, we have developed a computer vision-based helmet detection system that can detect if a rider has his helmet on in real-time. This paper presents a comprehensive review of the state-of-the-art computer vision-based helmet detection systems for two-wheelers. The review covers various aspects of helmet detection, including image pre-processing, feature extraction, and classification. The strengths and limitations of existing approaches are analyzed, while proposing potential directions for future research. The results demonstrate that computer vision-based helmet detection systems hold significant potential to
D, Bhavanash Rai
Method of Generating Real-Time Digital Customer Feedback Loop for Connected Vehicle Applications.2024-26-02581/16/2024
Modern cars are essentially computers on wheels in terms of the technology stack going into the car. Similarly, the customer experience is gone beyond the conventional touch and feel to more digital experiences around the car. While creating digital experiments around the car, getting honest customer feedback on the same is essential. Getting customer feedback in real-time and at scale is a challenge. Generally, this is limited to surveys and campaigns OEMs do to collect specific intent. In this paper, we propose a method of generating a real-time customer digital feedback loop. This method relay on real-time sourcing of data from openly available sources like Twitter, mobile app stores, etc. The real-time collected data is run through an analytics system. The analytics system contains four key stages, data preparation, sentiment segmentation, topic modelling and inference presentation. This paper in detail covers advanced machine learning techniques like natural language processing
Tidke, Pradeep, Venugopal, Vaisakh, Shukla, Manoj, Jali, Nagarjuna, Rathod, Chirag
Brake Pad Life Monitoring System using Machine learning2024-26-00321/16/2024
In vehicles, Brake Pad is an important part for controlled driving and safety. Brake Pads are made up of friction materials which wear with use. Excessive pad wear may lead to increase the brake response time and NVH issues. Currently it is not feasible for customers to check the amount of brake pad wear without the support from experienced personnel as the unit is not visible from outside. In general no feature is provided by Indian OEMs for brake pad life estimation and indication. This paper proposes a hybrid machine learning approach to predict brake pad remaining useful life. It comprises of 3 modules namely weight, temperature and wear module. Weight module is based on mathematical formulation using longitudinal vehicle dynamics to estimate the weight of the vehicle which is essentially required to calculate kinetic energy dissipated during braking. Temperature and wear module are deep neural network based machine learning modules. The technology is developed in a simplified
Iqbal, Shoaib, BHAMBRI, MIHIR, lahase, Rahul
Recurrent Neural Network Model for On-Board Estimation of the Side-Slip Angle in a Four-Wheel Drive and Steering Vehicle15-17-01-0003-TEST-REC9/23/2023
Abstract A valuable quantity for analyzing the lateral dynamics of road vehicles is the side-slip angle, that is, the angle between the vehicle’s longitudinal axis and its speed direction. A reliable real-time side-slip angle value enables several features, such as stability controls, identification of understeer and oversteer conditions, estimation of lateral forces during cornering, or tire grip and wear estimation. Since the direct measurement of this variable can only be done with complex and expensive devices, it is worth trying to estimate it through virtual sensors based on mathematical models. This article illustrates a methodology for real-time on-board estimation of the side-slip angle through a machine learning model (SSE—side-slip estimator). It exploits a recurrent neural network trained and tested via on-road experimental data acquisition. In particular, the machine learning model only uses input signals from a standard road car sensor configuration. The model
Giuliacci, Tiziano Alberto, Ballesio, Stefano, Fainello, Marco, Mair, Ulrich, King, Julian
Applications of Neural Networks to Metallic Flexor Geometry Optimization of Flat Wipers15-17-01-00029/9/2023
Abstract In recent years, demands of flat wipers have rapidly increased in the vehicle industry due to their simpler structure compared to the conventional wipers. Procedures for evaluating the appropriate metallic flexor geometry, which is one of the major components of the flat wiper, were proposed in the authors’ previous study. However, the computational cost of the aforementioned procedures seems to be unaffordable to the industry. The discrete Winkler model regarding the flexor as the Euler–Bernoulli beam is established as the mathematical model in this study to simulate a flexor compressed against a surface at various wiping angles. The deflection of the beam is solved using a finite difference method, and the calculated contact pressure distributions agree fairly with those based on the corresponding finite element model. Flexor designs are paired with various windshield surfaces to accumulate a sufficiently large simulation database based on the mathematical model. An
Chu, Yi-Tzu, Huang, Ting-Chuan, Liao, Kuo-Chi
Categorizing Simulation Models Using Convolutional Neural Networks2023-01-12176/26/2023
Engineers come across frequent problems solved using artificial intelligence. Tasks of neural networks are often considered solving problems in such an intelligent manner. Whether as an optimization problem or development tool, neural networks help engineers and create a more efficient approach. Furthermore, information is stored in a unique data structure, during a project cycle. Engineers often face a time-consuming search to get an understanding of the project itself. This process of understanding the data needs to be repeated by every engineer, if not further documented. E.g. searching for a past project with similar parameters can be a big task, regarding huge databases. Worker fluctuation can increase the cost of this process as well. To address this issue, neural networks make a significant contribution. The main aspect of this paper is to add meta data to the core files of the project. With such labels it is possible to get access to the content of the model files of an engine
Grbavac, Andrija, Angerbauer, Martin, Grill, Michael, Kulzer, André
Application of AI/ML in Hydrobush Tuning to Enhance Overall Value Proposition2025-01-01325/5/2023
In this intense market of the automotive industry, optimization of vehicle components for superior performance and customer satisfaction is paramount. Hydrobushes play an integral role in achieving this within vehicle suspension systems by absorbing vibrations and improving ride comfort. The methods traditionally used for tuning these components are not only time-consuming but also heavily reliant on extensive empirical testing. This paper explores the ever-growing field of artificial intelligence (AI) and machine learning (ML) in the hydrobush tuning process. Leveraging AI/ML algorithms such as random forest, artificial neural network, and logistic regression to analyse vast datasets, identify patterns, and predict optimal configurations more efficiently than conventional methods. The stated approach accelerates the tuning process while also enhancing the precision of the outcomes, leading to superior ride quality and durability. Thus, the integration of AI/ML techniques in hydrobush
Hazra, Sandip, Khan, Arkadip Amitava
Investigations on the Effect of Injection Pressure and Timing on Noise and Vibration Characteristics of a Biodiesel Fueled Genset Engine and Artificial Neural Network Architecture2022-28-049911/18/2022
In this present investigation, combustion, noise and vibration characteristics were studied in a direct injection compression ignition engine by blending of 20% Pongamia biodiesel with neat diesel. Experiments were conducted with different fuel injection pressures and injection timings. The brake thermal efficiency of the biodiesel blend found increased to 1.75% when the fuel was injected at 200 bar and 23° before top dead center. Higher engine vibration of 5.5% was observed in Y-axis at the fuel injection pressure of 250 bar with the injection timing of 25° bTDC when comparing to the standard operating condition of the engine. The combustion noise level found decreased to 4% when the fuel injection time increased from 21 to 23° bTDC. From Artificial Neural Network analysis, two metrics, such as Coefficient of correlation (R) and Mean Square Error (MSE), were found as 0.99 and 12.6, respectively. The developed artificial neural network model predicts and optimizes the parameters used
Selvan, V. Arul Mozhi, G, Arumugam
3D Printing Materials with Tunable Mechanical PropertiesTBMG-4691211/1/2022
Teaching Drones How to Navigate Real-World WeatherTBMG-464959/1/2022
Right now, drones are either flown under controlled conditions, with no wind, or are operated by humans using remote controls. However, for drones to autonomously perform necessary but quotidian tasks, such as delivering packages or airlifting injured drivers from a traffic accident, drones must be able to adapt to wind conditions in real time.
USE OF AI BASED VISUAL INSPCTION FOR AEROSPACE PARTS INSPECTION AND MANUFACTURING APPLICATION Govind Yadav Siva Bavisetti Honeywell Technology Solutions Bangalore, Karnataka, India2022-26-12255/26/2022
ABSTRACT Aircraft manufacturing processes typically include one or more steps where the product is visually inspected for defects. Typically, visual inspection is a highly manual process that can be time consuming and prone to errors. Over the years, rule-based visual inspection machines have also emerged. However, each approach has drawbacks. Manual inspection requires the presence of a person, an inspector who performs assessment of the entity under question and passes judgement on it according to some training or previous knowledge. No equipment is required except the naked eye of the trained inspector. Visual inspection errors in manufacturing take one of two forms — missing an existing defect or incorrectly identifying a defect that does not exist (false positive). Misses tend to occur much more frequently than false alarms). Misses can lead to loss in quality, while false positives can cause unnecessary production costs and overall wastage. This paper studies the limitations of
Yadav, Govind, Bavisetti, Siva
An Autonomous Ground Based Multi Sensor Data Fused TDR Prediction and Monitoring System2022-26-12275/26/2022
According to a study conducted by IATA, EASA and ICAO, runway excursions are common type of incidents that leads to major accidents. Runway excursion predominantly occur due to incorrect utilization of the touchdown zone (TDZ). A ground-based multi-sensor data fused touchdown region (TDR) prediction and monitoring system is proposed in this paper. This system autonomously analyses real-time environmental data with remote sensing capability around the runway site, gathers real-time information on flight critical parameters of approaching aircraft through an ADS-B Out data link, predicts ideal TDR over the runway TDZ for a particular aircraft, and estimates the expected contemporaneous TDR based on the monitored current approach profile. If the ideal and expected TDR do not overlap, there is a chance for an excursion. In such cases, the flight parameter that needs corrective action is determined and notified to the pilot. In order to analyze dynamic environmental data, mathematical
Nagonda, Ajay Nagarajan, Manjunath, Karthik
AI deep learning-based bird detection for UAM and aircraft2022-26-12285/26/2022
Abstract - Safety of flight during takeoff, flight and landing is one of the important considerations for airborne vehicles. Bird strike is defined as a collision between a bird and aircraft during takeoff, flight and landing of airborne vehicles. When bird strike happens, it can result in damage to aircraft structure and loss of thrust which impacts safety of flight. The nature of aircraft damage can differ depending on the size of the aircraft. For smaller aircrafts, it can result in penetration into flight deck windscreens or damage to control surfaces. In case of larger aircraft, it can result in hazardous effects because of engine ingestion. Many deep learning algorithms have evolved for object detection and classification in the recent years. You Only Look Once [YOLO] and single shot detector [SSD] are top two state of art object detection algorithms that are extensively used in real world applications. The Xilinx® Deep Learning Processor Unit (DPU) is a programmable engine
K, Mythri
A Framework for Teaching Safety Critical Control and Artificially Intelligent Systems to Undergrads2022-26-12235/26/2022
There is a need to teach systems to the undergrad and graduate levels in colleges. Efforts are being made by industry, academia, and professional societies to join hands to bridge the gap. The students require examples to get their “hands dirty” on an actual system. Many of the aerospace examples are not available in the open domain as they are considered proprietary. The world is going toward Artificial Intelligence. Though AI has been used in aerospace industry for optimization for quite some time, certification of AI systems in the safety critical control domain are still not very popular. Experiments on such system are required to clearly understand the testing and safety aspects of deploying such systems. This paper looks at an environmental control system which has enough complexity to merit an industry level attention but is again simple enough to be understood by the students. This is a physics model. Simulation runs from this physics model is used to train a Neural Network
Jeppu, Yogananda, Raman, Ramakrishnan
Navigating Deep Learning to Improve ADAS22AVEP04_044/1/2022
How edge computing drives real-time decision making in AI for smarter, safer automated vehicles. Automotive ‘big data’ is here. The immense scope of data generated by automated or ADAS-integrated vehicles spans the five SAE levels of autonomous driving, with reliance on high-resolution cameras, radar, lidar, ultrasonic sensors, GPS and other sensors for vehicles to see or perceive their surroundings. Ultimately, this sensory information - massive amounts of data - is used to navigate, avoid obstacles and read road markers necessary for safe driving. Artificial intelligence (AI) is at the heart of these operations, grounded in software algorithms and fueled by deep-learning training and deep-learning inference models that are essential to faultless performance. Enabling these vital and instantaneous processes requires AI algorithms to be trained and then deployed on-vehicle. It's a process that has developers tapping into both sophisticated software design and smart hardware strategies
Seetoo, Dustin, SintzUSER, Jeneane
Vehicle Feature Recognition Method Based on Image Semantic Segmentation2022-01-01653/29/2022
In the process of truck overload and over-limit detection, it is necessary to detect the characteristics of the vehicle's size, type, and wheel number, etc. In addition, in some vehicle visual load recognition systems, the vehicle load can be calculated by detecting the vibration frequency of specific parts of vehicle or the change in the length of the suspension during the vehicle's forward process. Therefore, it is very important to quickly and accurately identify vehicle features through the camera. This paper proposes a vehicle feature recognition method based on image semantic segmentation and Python, which can identify the length, height, number of wheels and vibration frequency at specific parts of the vehicle based on the vehicle driving video captured by the roadside camera. The process of vehicle recognition is as follows: First, build a convolutional neural network based on the image semantic segmentation model SegNet and use the data set to train it, and use the Python
Wang, Chu, Ding, Shousheng
The Virtual Boosted DISI Engine Model Development Based on Artificial Neural Networks2022-01-04563/29/2022
To efficiently reduce the required experimental data and improve the prediction accuracy, a virtual engine mod-el has been built by integrating an artificial neural network (ANN) system consisting of multiple subnets with the principle component analysis (PCA) method. The PCA method could preprocess the training data and leads to a more efficient training process. By coupling with 1-D GT power platform, this model has been adopted to predict the combustion phases (including CA10, CA50 and CA90), exhaust gas temperature, brake specific fuel consumption rate (be) and engine emissions such as un-burnt hydrocarbon (UBHC), NOx and CO. The results are then compared with the experimental data from around 5000 operating points of a boosted DISI engine running at universal performance map and conditions with various valve timing configurations. The mean abso-lute errors in combustion phases and exhausted gas temperature are less than 1.0 °CA and 10.1 K, respectively. The average relative errors
Chen, Ceyuan, Wu, Jian, Wei, Jingsi, Xu, Hanjun
Road Crossing Assistance Method Using Object Detection Based on Deep Learning 2022-01-01693/29/2022
This paper describes a method for assisting pedestrians to cross a road. As motorization develops, pedestrian protection techniques are becoming more and more important. Advanced driving assistance systems (ADAS) are improving rapidly to provide even greater safety. However, since the accident risk of pedestrians remains high, the development of an advanced walking assistance system for pedestrian protection may be an effective means of reducing pedestrian accidents. Crossing a road is one of the highest risk events, and is a complex phenomenon that consists of many dynamically changing elements such as vehicles, traffic signals, bicycles, and the like. A road crossing assistance system requires three items: real-time situational recognition, a robust decision-making function, and reliable information transmission. Edge devices equipped with autonomous systems are one means of achieving these requirements. Situational recognition when crossing a road must identify the pedestrian
Kawamura, Hiroaki, Shintani, Kohei, Mima, Hiroki, Taniguchi, Mashio
Databased modeling: An AI Toolchain for the development process of combustion engines2022-01-01883/29/2022
Predictive physical modeling is an established method used in the development process for any automotive component or system. While accurate predictions can be issued after tuning model parameters, long computation times can be expected depending on the complexity of the model at hand. As the requirements of the components/systems to be developed continuously increase, new optimization approaches are constantly being applied to solve multidimensional objectives and resulting conflicts optimally. Unfortunately, some of those approaches are deemed not feasible, as the computational times of the required single predictions using conventional simulation models are too high. Therefore, it is proposed to use data-based models such as trained neural networks instead of physical models to address this issue. Previous efforts have failed due to a weak database and the resulting poor predictive ability of data-based models. This paper introduces an AI Toolchain used for data-based modeling of
Milojević, Sasa, Rether, Dominik, Bodza, Sebasian, Grill, Michael, Angerbauer, Michael, Bargende, Michael, Cimniak, Valerian
Machine learning assisted analysis of heat transfer characteristics of a heavy duty natural gas engine2022-01-05723/29/2022
Heat transfer affects engine performance, efficiency, and emissions. Hence, it was essential to investigate the heat transfer characteristics after converting a compression ignition engine to a spark ignition and natural gas. The heat transfer for each engine speed and load was calculated based on Woschni’s correlation. As the number of experiments that provided engine response was limited, an artificial neural network approach was proposed to predict the engine heat transfer characteristics and aid in the analysis. The good agreement between the experimental results and data predicted by the machine learning algorithm validated the predictive capability of the artificial neural network model. The results of the data-driven analyses suggested that the heat transfer increased as the spark timing advanced. Moreover, increasing the engine speed reduced the heat flux form the combustion chamber to the engine coolant. Further, leaning the mixture lowered the heat transfer rate to the
Liu, Jinlong, Dumitrescu, Cosmin, Ulishney, Christopher
Construction of Personalized Driver Models based on LSTM using Driving Simulator.2022-01-09813/29/2022
The purpose of this study is to construct personalized driver models using Long Short-Term Memory (LSTM). The driver model is used for automated driving systems and driver assistance systems and is expected to control longitudinal and lateral directions of vehicles. For example, driver models including the individual driving characteristics adapt the control of systems, which can reduce discomfort and nuisance to drivers. The LSTM models in this study enable the time series data processing. In some previous studies, LSTM models have been used to investigate pedestrian behaviors. However, there are few studies of driver behavior models based on LSTM. In this study, we measured the data of the driver's driving operation using a Driving Simulator (DS). DS consists of a visual display and a motion display, of which motion functions consist of a 6-axis motion device and a turntable. DS has a cockpit using an actual body of a vehicle, and a visual information projected on a cylindrical
Hamada, Ayaka, Oikawa PhD, Shoko, Hirose, Toshiya
A prediction model of octane number loss based on neural network2022-01-01813/29/2022
The octane number is the most important indicator of motor petrol, and the petrol refining process is one of the important links in petrol production. The prediction of the octane loss number of petrol during the refining process is helpful to the improvement of petrol refining process and the processing of petrol. The traditional octane number prediction method relied on physical and chemical properties, and did not fully consider the high nonlinearity and strong coupling relationship of the petrol refining process.There is a lack of data-driven octane loss models. This paper studies the construction of the octane loss model in the petrol refining process. The main innovation is to use frequency statistics to summarize the operating range of data variables and preprocess the dataset, and use the variance selection method to filter out some invalid variables, then use the multi-dimensional association analysis to select 27 main operating variables in the next step. Finally, a neural
yang, Gang, Zhang, Sumin, Li, Shuai, Li, Wenju
Estimation of Surface Temperature Distributions Across an Array of Lithium-Ion Battery Cells Using a Long Short-Term Memory Neural Network2022-01-08453/29/2022
As electric vehicles are becoming increasingly popular and necessary for the future mobility needs of civilization, further effort is continually made to improve the efficiency, cost, and safety of the lithium-ion battery packs that power these vehicles. To facilitate these goals, this paper introduces one possible type of model to predict a distribution of surface temperatures for a lithium-ion battery pack: a long short-term memory (LSTM) neural network. The LSTM model is trained and validated with lithium-ion cells electrically connected to form a battery pack. Voltage, current, state of charge (SOC), and cell surface temperature from one array are used as inputs from a wide range of high and low temperature drive cycles. Additionally, a second model is considered where ambient temperature is added as an input to the LSTM model and the resulting predictions are compared to the original model. In summary the LSTM model can accurately characterize and predict a distribution of lithium
Campbell, Jeffrey, ZHU, DI, Cho, Gyouho
Adjoint-Based Model Tuning and Machine Learning Strategy for Turbulence Model Improvement2022-01-10793/29/2022
As turbulence modeling has become an indispensable approach to perform flow simulation in a wide range of industrial applications, how to improve the prediction accuracy has gained increasing attention during the past years. Of all the turbulence models, RANS is the most common choice for many OEMs due to its short turn-around time and strong robustness, however, the default setting of RANS is usually benchmarked through classical and well-studied engineering examples, not always suitable for resolving complex flows in specific applications. Many previous researches have suggested a small tuning in turbulence model coefficients could achieve higher accuracy on a variety of flow scenarios. Instead of adjusting parameters by trial and error from experience, this paper introduced a new data-driven approach of turbulence model recalibration using adjoint solver, based on Generalized k-ω (GEKO) model, one variant of RANS. In this approach, the model coefficients are tuned locally to match
Zhou, Hua, Ren, Chao
An Efficient Machine Learning Algorithm for Valve Fault Detection 2022-01-01853/29/2022
Multi-level Miller-cycle Dynamic Skip Fire (mDSF) is a combustion engine technology that improves fuel efficiency by deciding on each cylinder-event whether to skip (deactivate) the cylinder, fire with low (Miller) charge, or fire with a high (Power) charge. In an engine with two intake and two exhaust valves, skipping the cylinder is accomplished by deactivating all valves, while firing with a reduced charge is accomplished by deactivating one of the intake valves. This new ability to modulate the charge level introduces new failure modes. The first is a failure to reactivate the single intake valve, which results in a desired high-fire having the air intake of a low-fire; the second is a failure to deactivate the single intake valve, which results in a low-fire having the air intake of a high-fire. Reliably detecting these two faults has proven challenging for classical techniques that use measured MAP and/or crank angle acceleration to identify characteristic features of the
Serrano, Joe, Ortiz-Soto, Elliott, Chen, S Kevin, Chien, Li-Chun, Joshi, Abhishek
Items per page:
1 – 50 of 532