Browse Topic: People and personalities
The ISO TR 5469 Technical Report provides a framework to classify the AI/ML technology based on usage level and the properties and requirements to mitigate cyber and functional safety risks for the technology. This paper provides an overview of the approach used by ISO TR 5469 as well as an example of how one of the six ISO TR 5469 desirable properties (resilience to adversarial and intentional malicious input) can be analyzed for adversarial attacks. This paper will also show how a vehicle testbed can be used to provide a student with an AI model that can be used to simulate a non-targeted cyber security attack. The testbed can be used to simulate a poisoning attack where the student can manipulate a training data set to deceive the AI model during a simulated deployment.1 The University of Detroit Mercy (UDM) has developed Cyber-security Labs as a Service (CLaaS) to support teaching students how to understand and mitigate cyber security attacks. The UDM Vehicle Cyber Engineering (VCE
The research activity aims at defining specific Operational Design Domains (ODDs) representative of Italian traffic environments. The paper focuses on the human-machine interaction in Automated Driving (AD), with a focus on take-over scenarios. The study, part of the European/Italian project “Interaction of Humans with Level 4 AVs in an Italian Environment - HL4IT”, describes suitable methods to investigate the effect of the Take-Over Request (TOR) on the human driver’s psychophysiological response. The DriSMI dynamic driving simulator at Politecnico di Milano has been used to analyse three different take-over situations. Participants are required to regain control of the vehicle, after a take-over request, and to navigate through a urban, suburban and highway scenario. The psychophysiological characterization of the drivers, through psychological questionnaires and physiological measures, allows for analyzing human factors in automated vehicles interactions and for contributing to
Drivers present diverse landscapes with their distinct personalities, preferences, and driving habits influenced by many factors. Though drivers' behavior is highly variable, they can exhibit clear patterns that make sorting them into one category or another possible. Discrete segmentation provides an effective way to categorize and address the differences in driving style. The segmentation approach offers many benefits, including simplification, measurement, proven methodology, customization, and safety. Numerous studies have investigated driving style classification using real-world vehicle data. These studies employed various methods to identify and categorize distinct driving patterns, including naturalist differences in driving and field operational tests. This paper presents a novel hybrid approach for segmenting driver behavior based on their driving patterns. We leverage vehicle acceleration data to create granular driver segments by combining event and trip-based methodologies
Reproducing driving scenarios involving near-collisions and collisions in a simulator can be useful in the development and testing of autonomous vehicles, as it provides a safe environment to explore detailed vehicular behavior during these critical events. CARLA, an open-source driving simulator, has been widely used for reproducing driving scenarios. CARLA allows for both manual control and traffic manager control (the module that controls vehicles in autopilot manner in the simulation). However, current versions of CARLA are limited to setting the start and destination points for vehicles that are controlled by traffic manager, and are unable to replay precise waypoint paths that are collected from real-world collision and near-collision scenarios, due to the fact that the collision-free pathfinding modules are built into the system. This paper presents an extension to CARLA’s source code, enabling the replay of exact vehicle trajectories, irrespective of safety implications
Model-based developers are turning to DevOps principles and toolchains to increase engineering efficiency, improve model quality and to facilitate collaboration between large teams. Mature DevOps processes achieve these through automation. This paper demonstrates how integrating modern version control (Git) with collaborative development practices and automated quality enforcement can streamline workflows for large teams using Simulink. The focus is on enhancing model consistency, enabling team collaboration, and development speed.
This paper presents a new regression model-based method for accurate predictions of stiffness of different glass laminate constructions with a point-load bending test setup. Numerical FEA models have been developed and validated with experimental data, then used to provide training data required for the statistical model. The multi-variable regression method considered six input variables of total glass thickness, thickness ratio of glass plies as well as high-order terms. Highly asymmetrical, hybrid laminates combining a relatively thick soda-lime glass (SLG) ply joined with a relatively thin Corning® Gorilla® Glass (GG) ply were analyzed and compared to standard symmetrical SLG-SLG constructions or a monolithic SLG with the same total glass thickness. Both stiffness of the asymmetrical laminates and the improvement percentage over the standard symmetrical design can be predicted through the model with high precision.
In the automotive industry, there have been many efforts of late in using Machine Learning tools to aid crash virtual simulations and further decrease product development time and cost. As the simulation world grapples with how best to incorporate ML techniques, two main challenges are evident. There is the risk of giving flawed recommendations to the design engineer if the training data has some suspect data. In addition, the complexity of porting simulation data back and forth to a Machine Learning software can make the process cumbersome for the average CAE engineer to set up and execute a ML project. We would like to put forth a ML workflow/platform that a typical CAE engineer can use to create training data, train a PINN (Physics Informed Neural Network) ML model and use it to predict, optimize and even synthesize for any given crash problem. The key enabler is the use of an industry first data structure named mwplot that can store diverse types of training data - scalars, vectors
This research explores the use of salt gradient solar ponds (SGSPs) as an environmentally friendly and efficient method for thermal energy storage. The study focuses on the design, construction, and performance evaluation of SGSP systems integrated with reflectors, comparing their effectiveness against conventional SGSP setups without reflectors. Both experimental and numerical methods are employed to thoroughly assess the thermal behavior and energy efficiency of these systems. The findings reveal that the SGSP with reflectors (SGSP-R) achieves significantly higher temperatures across all three zones—Upper Convective Zone (UCZ), Non-Convective Zone (NCZ), and Lower Convective Zone (LCZ)—with recorded temperatures of 40.56°C, 54.2°C, and 63.1°C, respectively. These values represent an increase of 6.33%, 11.12%, and 14.26% over the temperatures observed in the conventional SGSP (SGSP-C). Furthermore, the energy efficiency improvements in the UCZ, NCZ, and LCZ for the SGSP-R are
Monitoring the rotor temperature of drive machines is crucial for the safety and performance of electric vehicles. However, due to the complex operating conditions of electric vehicles, the thermal parameters of vehicular induction machines (IMs) vary significantly and are difficult to identify accurately. This article first establishes a concise but effective thermal network for IMs and analyzes the influencing factors of thermal parameters. Then, a parameter identification network (PIN) with multiple parallel branches is constructed to learn the mapping relationship between electromechanical variables and thermal parameters. Afterward, temperature datasets for network training are built through bench testing. Finally, the effectiveness of identified parameters for rotor temperature estimation application is verified, demonstrating improved interpretability, generalization ability, and accuracy compared to an end-to-end neural network.
Items per page:
50
1 – 50 of 10198