Self-Driving Cars

In 2014, a future route was planned between Rotterdam and Vienna, where – in the coming years - they want to realize the first cross-Europe corridor of the cooperative, intelligent transport system. Even in everyday life, we hear about the increased use of autonomous vehicles now. We are now able to come across driverless metros from Budapest to Paris. The marketable variation of the different automated vehicles (such as buses, truck, cars) are being developed and tested. We’ve been witnessing the success and failure of some of the early models. According to this, the Europe-wide mobility will – with the networking of the various transportation systems and spreading of the individual and public transportation that is capable of autonomous operation - probably change the verticum of the whole society, which will create one of the essential pillars of the intelligent city.

Ciceró védjegye

A „Ciceró védjegye” kifejezés a védjegyjog szempontjából különleges jelentőséggel bír. A védjegy alapvető funkciói közé tartozik a megkülönböztetőképesség, az eredetjelzés és a fogyasztóvédelem. A Ciceró névhez kötött megjelölés ugyanakkor kulturális örökségi dimenzióval is rendelkezik: a latin retorikai hagyomány részeként a közösségi emlékezetben rögzült.

A jogi probléma kettős:

Monopólium tilalma – a jogrendnek biztosítania kell, hogy a közkinccsé vált kulturális jelek ne kerüljenek egyetlen piaci szereplő kizárólagos rendelkezése alá.

Megfelelő védelem – ugyanakkor indokolt a kereskedelmi visszaélések elleni fellépés, amikor a megjelölést olyan módon használják, amely sérti a kulturális identitást vagy megtéveszti a fogyasztókat.

A „Ciceró védjegye” így példát ad arra, hogyan kell a védjegyjogot a kulturális örökség megőrzésének szempontjával összeegyeztetni.

Purpose: This paper aims to explore the potential impact of autonomous vehicles (AVs) on urban planning, sustainable urban development, and tourism. Methodology: The paper is a conceptual study that reviews and synthesizes existing literature on AVs, urban planning, and tourism. It also uses case studies to illustrate the potential effects of AVs. Results: The widespread adoption of AVs is likely to have significant implications for urban planning, including changes in land use, infrastructure design, and transportation patterns. AVs may also contribute to sustainable urban development by reducing traffic congestion and air pollution. In the tourism sector, AVs could lead to spatial changes, new social inequalities, and changes in the overnight visitor economy. Theoretical contribution: The paper contributes to the understanding of the complex interplay between AVs, urban planning, and tourism. It highlights the need for urban planners and tourism stakeholders to consider the potential impact of AVs in their decision-making processes. Practical implications: The paper provides practical insights for urban planners, tourism stakeholders, and policymakers on how to prepare for and adapt to the widespread use of AVs. It also emphasizes the importance of considering the potential benefits and challenges of AVs in the context of specific cities and tourism destinations.

This study presents an intelligent robotic object grasping system using computer vision technique and deep reinforcement learning to enhance robotic manipulation. The proposed technique employs You Only Look Once (YOLOv3) for real-time object recognition and localisation, while the Soft Actor-Critic (SAC) system uses depth image information to determine the optimal gripping areas. By transforming the gripping point into a three-dimensional grasping posture, the robotic manipulator can then efficiently choose and arrange objects. The COCO dataset was utilised to increase YOLO's detection accuracy, and transfer learning sped up the training process. The performance evaluation of the proposed system revealed a mean Average Precision (mAP) of 91.2% for item detection and an 87.3% grasping success rate. 10-fold cross-validation verified the model's robustness and generalisability, demonstrating minimal variation in performance across test settings. Compared to traditional gripping approaches, the proposed strategy improved accuracy by 27% and execution efficiency by 35%. These findings demonstrate the YOLO-SAC framework's promise for practical robotic applications by providing a flexible and scalable approach to automated object handling in a range of settings.

Convolutional Neural Networks (CNNs) have transformed the area of deep learning by demonstrating exceptional abilities in a range of tasks, including object identification, picture identification, and natural language processing. Yet, study and investigation into the complex interactions between various architectural elements inside CNNs known as architectural synergy remain continuing. This study addresses the consequences of architectural synergy for improving model performance, scalability, and reliability, and examines how it enables deep learning using CNNs. Using an extensive examination of extant literature and real-world implementation cases, we clarify the processes that underlie architectural synergy and underscore its capacity to enhance the capabilities of CNN-powered models. We hope to further our knowledge of the fundamental ideas behind the effectiveness of CNNs in deep learning by illuminating this important area of neural network architecture learning tasks.

This project presents an automated face recognition and attendance monitoring system using deep learning and machine learning techniques. The system leverages a CNN ensemble for facial feature extraction and BERT embeddings for textual data. The model achieves high accuracy on the ORL dataset and provides visual insights using confusion matrices, ROC curves, t-SNE plots, and attendance timelines. The system is designed for deployment in real-world environments, offering reliable and interpretable results.

This comprehensive article explores the transformative impact of GitOps on modern software development and operations practices. GitOps, a methodology that leverages Git repositories as the single source of truth for infrastructure and application management, has emerged as a powerful approach to enhance consistency, traceability, and automation in operational workflows. The article delves into the fundamental principles of GitOps, examining its core components and the benefits it brings to organizational processes. It provides an in-depth analysis of popular GitOps tools such as ArgoCD and Flux, discussing their integration with CI/CD pipelines and comparing their features and capabilities. The challenges encountered during GitOps implementation are addressed, along with best practices for successful adoption, including declarative Infrastructure as Code, immutable deployments, and automated synchronization. The article also explores the impact of GitOps on organizational dynamics, highlighting improvements in deployment consistency, traceability, and collaboration between development and operations teams. Looking toward the future, the article discusses emerging trends and technologies that are likely to shape the evolution of GitOps, including AI integration, serverless architectures, and enhanced security practices. By offering a holistic view of GitOpsfrom its current state to future prospectsthis article serves as a valuable resource for organizations seeking to optimize their software delivery processes and embrace cutting-edge DevOps practices in an increasingly cloudnative world.

El desarrollo de los autos autónomos es un proyecto que ha estado presente en la agenda de las grandes compañías e ingenieros desde por lo menos 1920. En el año 1921, en Dayton, Ohio, ingenieros de Radio AirService presentaron en la base de pruebas de la fuerza aérea McCook el primer auto sin conductor y manejado por radio (Kröger, 2016). Las décadas siguientes lo sucedieron una serie de autos remotos mejorados, pero no fue hasta 1977 donde aparece el primer auto computarizado sin conductor, construido por Sukuba Mechanical Engineering Lab, en Japón. y con visión artificial, capaz de detectar las líneas de división de carriles. A medida que el procesamiento de imágenes, CPU, y demás componentes fueron mejorando, el sueño del auto autónomo se fue tornando realidad.
En 1995 se logró un hito importante cuando el NavLab 5 de CMU AVL completó el primer viaje autónomo de costa a costa de los Estados Unidos (Kanade, Thorpe, & Whittaker, 1986).
Si bien en las últimas décadas se ha avanzado mucho gracias a la incorporación de nuevos sensores, actuadores, incremento de poder de procesamiento, nuevos y mejorados algoritmos de inteligencia artificial, y experiencia ganada en trayectorias realizadas (Krafcik, 2018) de manera autónoma, no existe un auto 100% autónomo de vehículos de venta al público.

Autonomous or "self-driving" cars are vehicles that drive themselves without human supervision or input. Because of safety benefits that they are expected to bring, autonomous vehicles are likely to become more common. Notably, for the first time, people will share a physical environment with computer-controlled machines that can both direct their own activities and that have considerable range of movement. This represents a distinct change from our current context. Today people share physical spaces either with machines that have free range of movement, but are controlled by people (e.g. automobiles) or with machines that are controlled by computers, but highly constrained in their range of movement (e.g. elevators). The movements of today's machines are thus broadly predictable. The unrestricted, computer-directed movement of autonomous vehicles is an entirely novel phenomenon that may challenge certain unarticulated assumptions in our existing legal structure. Problematically, the movements of autonomous vehicles may be less predictable to the ordinary people who will share their physical environment-such as pedestrians-than the comparable movements of human-driven vehicles. Today, a great deal of physical harm that might otherwise occur is likely avoided through humanity's collective ability to predict the movements of other people. In anticipating the behavior of others, we employ what psychologists call a "theory of mind." Theory of mind cognitive mechanisms allow us to extrapolate from our own internal mental states in order to estimate what others are thinking or likely to do. These cognitive systems allow us to make instantaneous, unconscious judgments about the likely actions of people around us, and therefore, to keep ourselves safe in the driving context. However, the theory of mind mechanisms that allow us to accurately model the minds of other people and interpret their communicative signals of attention and

Het Kennisinstituut voor de Mobiliteit (KiM) beschrijft in “Chauffeur aan het stuur?” vier toekomstbeelden voor de zelfrijdende auto (ZRA). Deze beelden varieren zowel in de mate van automatisering (overal of met name op de snelweg) als in de mate van autodelen (nauwelijks of veel). Het KiM schat daarbij ook de maatschappelijke gevolgen. Als we de effecten voor specifieke groepen (OV, autodealers etc.) buiten beschouwing laten, valt op dat (veel) ZRA technologie goed is voor verkeersveiligheid, sociale inclusie en ruimtelijke ordening, maar negatief uitpakt voor milieu en leefbaarheid, fietsen (gezondheid) en files (de omvang van het verkeer groeit harder dan de wegcapaciteit). Alleen wanneer technologie gecombineerd wordt met autodelen vallen deze negatieve aspecten weg en kun je van een (volledig) succesvolle ontwikkeling spreken. Maar die combinatie is allerminst zeker en gedrag –hier is dat delen- is lastig te sturen. Hoe zorgen we ervoor dat de ZRA een commercieel en maatschapp...

Autonomous vehicles are complex systems that operate in dynamic environments, where automated driving seeks to control the coupled longitudinal and lateral vehicle dynamics to follow a certain behaviour. Model predictive control is one of the most promising tools for this type of application due to its optimal performance and ability to handle input and output constraints. This paper addresses autonomous driving by introducing an adaptive linear parameter varying model predictive controller (LPV-MPC), whose prediction model is adapted online by a neural network. Moreover, the controller's cost function is optimized by an improved Genetic Algorithm. The proposed controller is evaluated on a challenging track subject to variable wind disturbances.

Autonomous driving is a complex and highly dynamic process that ensures controlling the coupled longitudinal and lateral vehicle dynamics. Model predictive control, distinguished by its predictive feature, optimal performance, and ability to handle constraints, makes it one of the most promising tools for this type of control application. The content of this article handles the problem of autonomous driving by proposing an adaptive linear parameter varying model predictive controller (LPV-MPC), where the controller's prediction model is adaptive by means of a recurrent neural network. The proposed LPV-MPC is further optimised by a hybrid Genetic and Particle Swarm Optimization Algorithm (GA-PSO). The developed controller is tested and evaluated on a challenging track under variable wind disturbance.

Reinforcement learning (RL) is widely used in autonomous driving tasks and training RL models typically involves in a multi-step process: pre-training RL models on simulators, uploading the pre-trained model to real-life robots, and fine-tuning the weight parameters on robot vehicles. This sequential process is extremely time-consuming and more importantly, knowledge from the fine-tuned model stays local and can not be re-used or leveraged collaboratively. To tackle this problem, we present an online federated RL transfer process for real-time knowledge extraction where all the participant agents make corresponding actions with the knowledge learned by others, even when they are acting in very different environments. To validate the effectiveness of the proposed approach, we constructed a real-life collision avoidance system with Microsoft Airsim simulator and NVIDIA JetsonTX2 car agents, which cooperatively learn from scratch to avoid collisions in indoor environment with obstacle objects. We demonstrate that with the proposed framework, the simulator car agents can transfer knowledge to the RC cars in real-time, with 27% increase in the average distance with obstacles and 42% decrease in the collision counts.

Reinforcement learning (RL) is widely used in autonomous driving tasks and training RL models typically involves in a multi-step process: pre-training RL models on simulators, uploading the pre-trained model to real-life robots, and fine-tuning the weight parameters on robot vehicles. This sequential process is extremely time-consuming and more importantly, knowledge from the fine-tuned model stays local and can not be re-used or leveraged collaboratively. To tackle this problem, we present an online federated RL transfer process for real-time knowledge extraction where all the participant agents make corresponding actions with the knowledge learned by others, even when they are acting in very different environments. To validate the effectiveness of the proposed approach, we constructed a real-life collision avoidance system with Microsoft Airsim simulator and NVIDIA JetsonTX2 car agents, which cooperatively learn from scratch to avoid collisions in indoor environment with obstacle objects. We demonstrate that with the proposed framework, the simulator car agents can transfer knowledge to the RC cars in real-time, with 27% increase in the average distance with obstacles and 42% decrease in the collision counts.

Drones with obstacle avoidance capabilities have attracted much attention from researchers recently. They typically adopt either supervised learning or reinforcement learning (RL) for training their networks. The drawback of supervised learning is that labeling of the massive dataset is laborious and time-consuming, whereas RL aims to overcome such a problem by letting an agent learn with the data from its environment. The present study aims to utilize diverse RL within two categories: (1) discrete action space and (2) continuous action space. The former has the advantage in optimization for vision datasets, but such actions can lead to unnatural behavior. For the latter, we propose a U-net based segmentation model with an actor-critic network. Performance is compared between these RL algorithms with three different environments such as the woodland, block world, and the arena world, as well as racing with human pilots. Results suggest that our best continuous algorithm easily outpe...

In this paper, a cooperative decision-making is presented, which is suitable for intention-aware automated vehicle functions. With an increasing number of highly automated and autonomous vehicles on public roads, trust is a very important issue regarding their acceptance in our society. The most challenging scenarios arise at low driving speeds of these highly automated and autonomous vehicles, where interactions with vulnerable road users likely occur. Such interactions must be addressed by the automation of the vehicle. The novelties of this paper are the adaptation of a general cooperative and shared control framework to this novel use case and the application of an explicit prediction model of the pedestrian. An extensive comparison with state-of-the-art algorithms is provided in a simplified test environment. The results show the superiority of the proposed model-based algorithm compared to state-of-the-art solutions and its suitability for real-world applications due to its real-time capability.

Most autonomous vehicle (AV) systems today operate like lone wolves-high-tech, self-reliant, and isolated. A single vehicle collects data, interprets its surroundings, and makes decisions independently. It may use advanced tools like LiDAR, radar, or neural networks, but it ultimately remains a solitary agent in a shared environment. This isn't just a coordination issue; it's a fundamental misinterpretation of what autonomy could be. Valence rethinks this paradigm. Rather than viewing each vehicle as an isolated intelligence, it treats them as interconnected neurons within a larger synthetic brain. Each vehicle shares data and intentions with others, forming a dynamic, distributed cognitive ecosystem. This paper explores the system's architecture, learning framework, philosophical rationale, and the tangible advantages of this collaborative model. The Future of AVs: A Shift That's Already Happening Autonomous vehicles are no longer a distant future-they are an emerging reality driven by urban congestion, sustainability goals, accident reduction, and technological readiness. But as AVs proliferate, the real bottleneck is not perception or motion planning, but interaction. Effective V2V (vehicle-to-vehicle) communication is becoming as vital as steering or braking. Systems like Valence, which support data exchange and shared decision-making, represent the connective tissue for next-generation mobility. Without this collaboration, even the most intelligent vehicles risk becoming dysfunctional in crowded, real-world traffic.

To plan safe trajectories in urban environments, autonomous vehicles must be able to quickly assess the future intentions of dynamic agents. Pedestrians are particularly challenging to model, as their motion patterns are often uncertain and/or unknown a priori. This paper presents a novel changepoint detection and clustering algorithm that, when coupled with offline unsupervised learning of a Gaussian process mixture model (DPGP), enables quick detection of changes in intent and online learning of motion patterns not seen in prior training data. The resulting long-term movement predictions demonstrate improved accuracy relative to offline learning alone, in terms of both intent and trajectory prediction. By embedding these predictions within a chanceconstrained motion planner, trajectories which are probabilistically safe to pedestrian motions can be identified in real-time. Hardware experiments demonstrate that this approach can accurately predict pedestrian motion patterns from onboard sensor/perception data and facilitate robust navigation within a dynamic environment.

This article explores a 1923 traffic accident in Triq tal-Imrieħel, Birkirkara, through the lens of a votive painting (ex-voto) donated to the Sanctuary of Our Lady of Sorrows (Madonna tal-Ħerba). The case involves Francesco Briffa, who suffered a serious leg injury after being struck by a car, and who later offered the painting in gratitude for what he believed was divine intercession. Drawing on archival police records from the National Archives of Malta, the study compares the official incident report—mistakenly listing the victim as Giuseppe Briffa—with the visual and textual testimony provided by the ex-voto. It examines the accuracy of the painting in depicting the injury and the accident scene, and discusses the devotional and historical value of such votive offerings. This study sheds light on how ex-voto art preserves forgotten or misrecorded episodes of personal and communal history, especially in the early era of motor vehicle incidents in Malta.

Introduction: Corridor selection plays a critical role in determining the optimal locations for deploying autonomous vehicles. With the rapid advancement of self-driving technology, identifying suitable corridors becomes paramount to maximize the efficiency, safety, and overall performance of autonomous vehicle systems. The process involves analyzing various factors such as traffic patterns, infrastructure conditions, regulatory frameworks, and user preferences. This paper explores the significance of corridor selection and highlights the key considerations involved, shedding light on the crucial decision-making process for locating autonomous vehicles in the most favorable corridors. The research on corridor selection for locating autonomous vehicles holds significant importance in the field of transportation and autonomous systems. It addresses the need to identify the most suitable corridors for deploying self-driving vehicles, taking into account factors such as traffic flow, infrastructure conditions, and regulatory aspects. The outcomes of this research can inform policymakers, urban planners, and transportation companies in making informed decisions regarding the placement of autonomous vehicles, ultimately leading to optimized traffic management, enhanced mobility, and improved safety on roadways. The methodology for studying the supplier selection process in the dairy industry will involve a combination of qualitative and quantitative approaches. Firstly, a comprehensive literature review will be conducted to gather insights into the existing supplier selection criteria and methodologies used in the industry. This will be followed by qualitative interviews and surveys with key stakeholders such as dairy company managers, procurement professionals, and suppliers to gather their perspectives on the supplier selection process. To find recurring themes and patterns in the acquired data, qualitative data analysis approaches will be used. Additionally, using statistical techniques and models like the Analytic Hierarchy Process (AHP) and TOPSIS, quantitative analysis will be done to assess the significance and effectiveness of various supplier selection criteria. To provide a comprehensive understanding of the supplier selection procedure in the dairy business, the results from both qualitative and quantitative assessments will be merged. road space availability, separation from car traffic, keeping pedestrians apart, number of travellers, possibility for revenue. road space availability, separation from car traffic, keeping pedestrians apart, number of travellers, Income potential from the result the availability of road has secured the first place and segregation from pedestrians stays at the least. availability of road plays an important role in building corridor selection for locating autonomous vehicles.

Implementing artificial neural networks is commonly achieved via high-level programming languages like Python and easy-to-use deep learning libraries like Keras. These software libraries come pre-loaded with a variety of network architectures, provide autodifferentiation, and support GPUs for fast and efficient computation. As a result, a deep learning practitioner will favor training a neural network model in Python, where these tools are readily available. However, many large-scale scientific computation projects are written in Fortran, making it difficult to integrate with modern deep learning methods. To alleviate this problem, we introduce a software library, the Fortran-Keras Bridge (FKB). This two-way bridge connects environments where deep learning resources are plentiful, with those where they are scarce. The paper describes several unique features offered by FKB, such as customizable layers, loss functions, and network ensembles. The paper concludes with a case study that applies FKB to address open questions about the robustness of an experimental approach to global climate simulation, in which subgrid physics are outsourced to deep neural network emulators. In this context, FKB enables a hyperparameter search of one hundred plus candidate models of subgrid cloud and radiation physics, initially implemented in Keras, to be transferred and used in Fortran. Such a process allows the model's emergent behavior to be assessed, i.e. when fit imperfections are coupled to explicit planetary-scale fluid dynamics. The results reveal a previously unrecognized strong relationship between offline validation error and online performance, in which the choice of optimizer proves unexpectedly critical. This in turn reveals many new neural network architectures that produce considerable improvements in climate model stability including some with reduced error, for an especially challenging training dataset.

The urban environment is increasingly engaging with artificial intelligence, a focus on the automation of urban processes, whether it be singular artefacts or city-wide systems. The impact of such technological innovation on the social dynamics of the urban environment is an ever changing and multi-faceted field of research. In this paper, the space and time defined by the autonomous vehicle is used as a window to view the way in which a shift in urban transport dynamics can impact the temporal experience of an individual. Using the finite window of time created by an autonomous vehicle, a theoretical framework is put forward that seeks to show how contrasting narratives exist regarding the experience of the window of time within the autonomous vehicle. By taking a theoretical approach informed by social theory, the dissolution of barriers between separate spheres of life is explored to highlight the increased commodification of time. In focusing on both the space and time created by the autonomous vehicle this approach seeks to highlight how artificial intelligence can provide a contemporary space in the urban environment while also opening a new window of time. The cognitive dissonance observed when comparing the narratives of autonomous vehicle stakeholders and the historical shift in time use leads to a belief that technology makes the user more free in terms of time. With this paper the autonomous vehicle is shown to be an ideal space and time to view the way in which the use of such technology does not increase free-time, but further dissolves the boundaries between what is and what is not work-time.

Automotive manufacturers are competing to be the first to introduce customer-ready autonomous vehicles. Some manufacturers are claiming to launch their first self-driving cars as early as 2020. Which all sounds very good and futuristic; however, the question arises, are customers even ready to adopt this new technological advancement? Therefore, this pilot study is aimed at finding out the answer to this question in the Austrian market. This study discovers the standpoint of Austrian consumers concerning the acceptance of self-driving cars for daily usage and gives an overview of the current point of view regarding autonomous vehicles (AVs). The data for this study was collected using an online, user-friendly, Likert scale survey. The collected data were processed and analyzed for empirical significance in SPSS using Spearman’s rank correlation and the Mann–Whitney U test supported by descriptive analysis. The results of the study indicate that Austrian consumers are well aware of a...

In the last couple of decades, there has been an unparalleled growth in number of people who can afford motorized vehicles. This is increasing the number of vehicles on roads at an alarming rate and existing infrastructure and conventional methods of traffic management are becoming inefficient both on highways and in urban areas. It is very important that our highways are up and running 24/7 as they not only provide a passage for human beings to move from one place to another, but also are the most important mode for intercity or international transfer of goods. There is an utter need of adapting the new world order, where daily processes are driven with the help of innovative technologies. It is highly likely that technological advancements like autonomous or connected vehicles, big data and the Internet of things can provide highway operators with a solution that might resolve unforeseeable challenges. This investigative exploratory research identifies and highlights the impact of...

A new method for detection of on-road vehicles based on color intensity segregation is proposed. This method has two steps. Firstly, details such as pavements or lanes in the image frame are utilized to extract the region of interest. Secondly, a new filter is proposed that utilizes the intensity information to filter the illumination variations, shadows and cluttered backgrounds from the extracted region of interest and detect the vehicles subsequently. The proposed method is evaluated on videos from KITTI vision benchmark suite and on our videos recorded during cloudy and rainy days with variable resolution. The experimental results demonstrate the effectiveness of the proposed vehicle detection method by achieving 90% detection rate and also reduced computational load on hardware, making it suitable for real-time applications.

Artificial Intelligence (AI) is transforming the world—reshaping industries, en- abling smarter systems, and driving innovation across every field. At the heart of this transformation lies a set of critical skills: the ability to write efficient programs, understand fundamental algorithms, and manipulate data with pre- cision. This book, Programming and Algorithms for AI, is written to equip learners with those essential capabilities.
My journey into this domain began over two decades ago with a degree in Applied Mathematics. Early roles in statistical data analysis gave me a deep appreciation for the power of numbers, logic, and structured thinking. As my career evolved through data mining and into data science, I discovered firsthand how foundational programming and algorithmic thinking are to building real- world AI solutions. It wasn’t just about knowing which model to apply—but understanding how to prepare the data, write the code, and optimize the pro- cess.
This book is my response to a growing need: a practical, beginner-friendly guide that bridges programming fundamentals and algorithmic thinking, all within the context of AI. Each chapter is short and to the point—designed to teach concepts clearly, reinforce them with examples, and demonstrate their relevance to AI applications. The goal is not to cover every corner of the subject, but to help you develop a strong foundation and the confidence to build on it.

The Reality Verification Problem: Mathematical Impossibility of Complete Self-Validation in Autonomous Systems
We introduce the Reality Verification Problem (RVP), a fundamental mathematical framework that proves complete self-validation is impossible for any sufficiently complex computational system. Our central theorem demonstrates that for any system S modeling reality, the computational requirements V(S) for verifying S's accuracy always exceed the system's own computational capacity Cap(S), establishing V(S) > Cap(S) as a universal constraint. This impossibility result has immediate implications for autonomous vehicles, AI safety protocols, financial trading systems, and cybersecurity frameworks requiring self-auditing capabilities.
We define the verification function V(S) and prove fundamental limits using diagonalization arguments analogous to Gödel's incompleteness theorems. Despite impossibility of complete verification, we develop practical approximation algorithms achieving (1-δ)-confident verification with polynomial overhead. Our framework provides the first rigorous mathematical foundation for understanding why autonomous systems cannot fully validate themselves and establishes optimal strategies for bounded verification within computational constraints.
Experimental validation demonstrates practical applications across critical domains: autonomous vehicle safety protocols achieving 85-95% verification confidence, AI alignment verification frameworks for AGI safety, financial algorithm self-auditing systems improving risk-adjusted returns by 15-25%, and cybersecurity self-monitoring platforms detecting threats with polynomial-time verification bounds. The impossibility results reveal fundamental trade-offs between system capability and verification requirements, providing essential mathematical foundations for safe deployment of autonomous systems.
This work establishes verification theory as a critical component of AI safety, autonomous systems engineering, and computational epistemology. As systems become increasingly autonomous, understanding mathematical limits of self-knowledge becomes essential for designing safe, reliable systems that acknowledge their own verification constraints. The RVP framework enables optimization of verification strategies within fundamental computational limits, supporting safer advancement of autonomous technologies across industries.
Keywords: autonomous systems, AI safety, self-verification, computational limits, system validation, verification theory, autonomous vehicles, algorithmic trading, cybersecurity, mathematical impossibility

The automotive industry has been focused on developing autonomous driving technology, which has been a subject of research for many years. Recent advancements in Convolutional Neural Networks (CNN) have shown remarkable performance in computer vision tasks, including autonomous driving systems. Behavioral cloning is a widely adopted technique in autonomous vehicle systems, that involves training a model to replicate the driving behavior of human drivers. This paper presents research on implementing behavioral cloning using a CNN architecture for an autonomous vehicle system in the Udacity self-driving car simulator environment. Our approach employs a customdesigned CNN Model to learn driving behavior from a dataset of images captured from the simulator, and we have applied various data augmentation techniques to the dataset to enhance the model's performance. This model has a total of 27 layers, including convolutional layers, normalization layers, activation layers, dropout layers, flatten layers, and dense layers. We evaluate the model's performance on a metric, called mean square loss error and demonstrate that our approach outperforms previous works in this domain. The results of this research underscore the effectiveness of using behavioral cloning and data augmentation techniques for autonomous vehicle systems.

This article explores the acceptance of autonomous systems in corporate management through the contribution of anthropomorphism. By ascribing human-like qualities to machines, anthropomorphism can make or break trust and acceptance of these systems by human workers. Our research reveals that although anthropomorphized AI creates a sense of reliability and familiarity, too many human-like qualities create discomfort, the so-called "uncanny valley" effect. This study delves into the implications of such dynamics on corporate governance, with a focus on the need for ethical models and openness in the utilization of AI systems. Through an exhaustive examination of case studies as well as empirical evidence, we determine the most crucial factors affecting employee acceptance as trustworthiness, psychological preparedness, and perceived advantages from AI adoption. The study adds to knowledge about how anthropomorphism impacts the relationship between human workers and autonomous systems, eventually offering frameworks for successful regulation and control of AI technologies within business contexts.

The rapid advancements in Artificial Intelligence (AI) are fundamentally transforming the landscape of autonomous vehicles and modern transportation systems. AIdriven technologies such as machine learning (ML), deep learning (DL), computer vision (CV), and natural language processing (NLP) have empowered vehicles to perceive, interpret, and navigate complex and dynamic environments with minimal or no human input. This transformative shift enables vehicles to make real-time decisions based on vast amounts of sensory data, improving both the safety and efficiency of transportation systems. Autonomous vehicles, powered by cutting-edge AI algorithms, are revolutionizing road safety by significantly reducing accidents caused by human error. AI systems enable real-time recognition of road conditions, pedestrian movement, traffic signals, and obstacles, allowing for rapid and precise decision-making that enhances collision avoidance and pedestrian safety. In addition to safety, AI contributes to traffic efficiency by optimizing traffic flow, reducing congestion, and enabling more efficient route planning, all of which reduce overall travel time and energy consumption. Furthermore, the integration of AI in transportation is altering urban mobility patterns by introducing autonomous taxis, delivery drones, and shared transportation services, creating a more interconnected and flexible transportation network. This paper delves into the transformative impact of AI on the design and functionality of autonomous vehicles and their integration within broader transportation ecosystems. A comprehensive literature survey highlights critical breakthroughs, including advances in sensor fusion, path planning, and decision-making algorithms, while also examining key technological milestones that have propelled the development of autonomous driving systems. However, despite these remarkable advancements, the adoption of fully autonomous vehicles faces persistent challenges. These include safety concerns, particularly in complex, unstructured environments, ethical considerations around decision-making in life-threatening situations, regulatory frameworks that are struggling to keep pace with the technology, and cybersecurity risks that threaten the safety and integrity of AI-driven systems. The working principles behind autonomous vehicles-such as the fusion of data from various sensors like LiDAR, radar, and cameras-are essential to creating a cohesive understanding of the vehicle's surroundings. Path planning algorithms and decision-making processes are crucial for ensuring that the vehicle can navigate efficiently and safely. This paper outlines these principles, providing insights into the current capabilities and limitations

In Sri Lanka, like many other countries, road signs play a crucial role in guiding traffic and ensuring road safety. In this research project, I propose a deep learning-based road signal detection and recognition system tailored specifically for Sri Lankan road conditions. My approach involves utilizing convolutional neural networks (CNNs) for image recognition using a comprehensive dataset of Sri Lankan Road signs, leveraging their ability to extract intricate features from visual data. We will use a moving vehicle with a camera installed to test and execute our system, mimicking actual driving situations. The vehicle's onboard technologies will use the trained model to enable real-time detection and recognition of road signs. We want to test the system extensively on various road types and climatic circumstances throughout Sri Lanka in order to assess its accuracy and resilience. A dependable and effective road signal detection and recognition system designed especially for Sri Lankan Road conditions is one of the research's anticipated results. Autonomous vehicles can operate securely and effectively, improving overall road safety and traffic efficiency in Sri Lanka, by precisely recognizing and interpreting road signs in real-time.

This article focuses on the economic potential and the consequences on the regulatory context of Shared Autonomous Vehicles (SAV) used in a regional public transportation system. Based on an experimental case study two on-demand scenarios were developed for the Swiss rural area of the T€ oss Valley. Scenario 1 replaces the current public transportation by SAVs; scenario 2 operates with a SAV fleet instead of buses and integrates the regional railway. Data sources are an overall traffic model and the current business figures of the public transportation. The results suggest that scenario 2 is, out of an economic and traffic system view, an attractive solution compared to line-bound traditional bus and train systems. In both scenarios, a cost-covering service may be possible due to an increase in productivity and demand. Regarding the present regulatory context of the Swiss public transportation system, we propose to change the system of call for tenders for single public transportation lines towards a call for tender for entire regions. This paper contributes to the scholarship discussion on the role of the final provider of new services and which adaptions of current regulations have to be targeted in the future.

Recently, machine learning has been very useful in solving diverse tasks with drones, such as autonomous navigation, visual surveillance, communication, disaster management, and agriculture. Among these machine learning, two representative paradigms have been widely utilized in such applications: supervised learning and reinforcement learning. Researchers prefer to use supervised learning, mostly based on convolutional neural networks, because of its robustness and ease of use but yet data labeling is laborious and time-consuming. On the other hand, when traditional reinforcement learning is combined with the deep neural network, it can be a very powerful tool to solve high-dimensional input problems such as image and video. Along with the fast development of reinforcement learning, many researchers utilize reinforcement learning in drone applications, and it often outperforms supervised learning. However, it usually requires the agent to explore the environment on a trial-and-error basis which is high cost and unrealistic in the real environment. Recent advances in simulated environments can allow an agent to learn by itself to overcome these drawbacks, although the gap between the real environment and the simulator has to be minimized in the end. In this sense, a realistic and reliable simulator is essential for reinforcement learning training. This paper investigates various drone simulators that work with diverse reinforcement learning architectures. The characteristics of the reinforcement learning-based drone simulators are analyzed and compared for the researchers who would like to employ them for their projects. Finally, we shed light on some challenges and potential directions for future drone simulators.

Traffic sign recognition (TSR) considered as a challenging subject in image processing for many years. Nowadays, after achievements in processing power of processors and easily accessible datasets, many researches has been done by using convolutional neural networks (CNN) In many applications including TSR. CNN is a popular deep learning method that has a reasonable functionality in image classification and pattern recognition. Important factors in performance of a CNN can be written as follows: accuracy, efficiency and the precision. Therefore, in this paper we try to achieve better results in these important factors. In this particular usage, we must consider complexity and processing time of whole procedure to be in an acceptable range. As will be shown in the following sections, because of wavelet characteristics, we will use wavelet in CNN to improve performance of it in two parts of its structure: In first step, we use wavelet convolutional neural network (wCNN) instead of CNN. Therefore, wavelet transform function is replaced in the convolutional layers of CNN. In next step, wavelet convolutional wavelet neural network (wCwNN) is suggested, so that fully connected neural network (FCNN) of wCNN and CNN is changed by wavelet neural network (wNN). So, we can compare results of these methods (wCNN and wCwNN) with CNN. We can obviously see that, accuracy results is about 4 percent better than CNN and the mean square error and the rate of error are decreased. These results achieved by increasing the calculations of the CNN algorithm.

The current paradigm for advancing autonomous driving functionalities on public roads predominantly operates within the tangible realm. Autonomous vehicles (AVs) are equipped with an array of sensors, including Lidar, cameras, radar, GPS, and on-boardunit (OBU) modems, each actively collecting data about the surrounding environment. This data is then processed through sophisticated algorithms, enhancing the vehicle's situational awareness. In our pursuit of refining autonomous driving capabilities, we leverage the AVTSC Simulator as our chosen Autonomous Vehicle (AV) simulation environment. Notably, this simulator stands out for its ability to replicate realistic urban settings and seamlessly integrate with widely used platforms such as Apollo and ROS. By harnessing the capabilities of this simulation environment, we aim to bridge the gap between real-world testing and simulated scenarios. This abstract encapsulates the essence of our strategy, emphasizing the integration of advanced sensor technologies and simulation environments for the development of autonomous driving features. Preliminary results indicate a promising 94% success rate for the proposed methodology, showcasing the efficacy of our approach in enhancing the robustness and reliability of autonomous systems.

This paper introduces a deep learning-based system for the detection of potholes on road surfaces using image classification techniques. The proposed approach employs a Convolutional Neural Network (CNN) to identify and classify images containing potholes from those without. A custom-designed graphical user interface (GUI) allows users to upload road images for instant analysis and detection. The system is trained on a publicly available dataset of road images, covering various pothole sizes and surface conditions. Designed for offline deployment, the system provides an efficient and accessible solution for road maintenance departments and municipal bodies, enabling timely and accurate road condition assessments.

This paper illustrates the MIR (Mobile Intelligent Robotics) Vehicle: a feasible option of transforming an electric ride-on-car into a modular Graphics Processing Unit (GPU) powered autonomous platform equipped with the capability that supports test and deployment of various intelligent autonomous vehicles algorithms. To use a platform for research, two components must be provided: perception and control. The sensors such as incremental encoders, an Inertial Measurement Unit (IMU), a camera, and a LIght Detection And Ranging (LIDAR) must be able to be installed on the platform to add the capability of environmental perception. A microcontroller-powered control box is designed to properly respond to the environmental changes by regulating drive and steering motors. This drive-by-wire capability is controlled by a GPU powered laptop computer where high-level perception algorithms are processed and complex actions are generated by various methods including behavior cloning using deep n...

autonomiset ajoneuvot, kuljetus-ja liikenneala, innovaation diffuusio, teknologian hyväksyntä, ihmisen ja koneen välinen suhde Tämä on deduktiivinen, selittävä tutkimus, joka käsittelee autonomisten ajoneuvojen (AV) käyttöönottohalukkuutta ja innovaatioiden diffuusiota. Tutkimuksen tarkoituksena oli selvittää esiintyykö kuluttajien keskuudessa riittävästi hyväksyntää ja käyttöhalukkuutta AV-teknologiaa kohtaan, jotta sen diffuusioprosessi voi käynnistyä. Kirjallisuuskatsaus käsitteli innovaatioiden käyttöönoton ja hyväksyttävyyden teoriaa ja aiempia tutkimuksia AV-teknologian vaikutuksista ja hyväksyttävyydestä. Empiirinen tutkimus toteutettiin kvantitatiivisin menetelmin. Kuluttajien AV-teknologian käyttöhalukkuutta mitattiin 300 vastaajan kyselyllä, jossa hyväksyntää ennustavat muuttujat oli valittu Car technology acceptance model eli CTAM-mallista ja innovaatioiden diffuusion teoriasta. Käyttöaikomuksia analysoitiin usean selittäjän lineaarisella regressiomallilla ja maksuhalukkuutta logistisella regressiomallilla. Tulosten mukaan vastaajilla on jokseenkin myönteinen näkemys AV-teknologiasta. Selkeä enemmistö piti AV-autoja turvallisempina ja parempina kuljettajina kuin ihmisisten ohjaamia autoja. 67.7 prosenttia vastaajista ilmaisi halukkuutta ottaa kyydin AVautossa, kun taas keskimääräinen maksuhalukkuus täyden itseohjautuvuuden mahdollistavasta lisävarusteesta oli 3 592 euroa. Turvallisuus, yhteensopivuus yhteiskunnan tarpeiden kanssa ja suhteelliset edut muihin liikennemuotoihin nähden vaikuttivat vahvimmin käyttöaikomuksiin ja maksuhalukkuuteen, kun taas demografisilla muuttujilla oli vähäisempi merkitys. Koska harvalla vastaajalla oli aiempia kokemuksia AV-teknologiasta, voidaan olettaa, että AV-teknologian hyväksyttävyys muuttuu, kun se tulee laajemmin kuluttajien saataville. may my thesis end up being ambitious or not, I was going to be in for a ride. That ride has been filled with discovery, joy, accomplishments and friendships, both in and out the class room. There will be so much to miss once I leave the campus for the last time as a student whether it is the sauna evenings in PK5, the 7th building gym or taking a course not for the credit points, but just for the sake of learning something useful. LUT is an excellent university and I am glad I chose to apply there. Unfortunately, my university years were also riddled with hardships everyone goes through as a natural part of life. While I possibly experienced a little more than most people do at one time, I am grateful for the folks who supported me through thick and thin. I want to thank them now. Starting with my family and relatives, to whom I can rely on. Anna, who always has the right words to say when I need to hear them. Milo, on whose couch I must have written more than half of this thesis paper. My supervisors Kaisu Puumalainen and Maija Hujala, who provided their insightful guidance throughout this process. Most importantly, I want to give a huge thank you to everyone who participated in my survey by answering and spreading it forward. Without you this thesis would be nothing but an empty husk. Thank you all dearly,

Autonomous vehicles have the potential to transform lives by providing transportation to a wider range of users. However, with this new method of transportation, user acceptance and comfort are critical for widespread adoption. This exploratory study aims to investigate what makes passengers uncomfortable in existing vehicles to inform the design of future autonomous vehicles. In order to predict what may impact user acceptance for a diverse rider population for future autonomous vehicles, it is important to understand what makes a broad range of passengers uncomfortable today. In this study, interviews were conducted for a total of 75 participants from three diverse groups, including 20 automotive engineering graduate students who are building an autonomous concept vehicle, 21 non-technical adults, and 34 senior citizens. The results revealed both topics which made different groups of passengers uncomfortable as well as how these varied between the groups. The leading contributors to the highest discomfort for all groups were being a passenger in situations with a distracted driver, being in a vehicle that is following too closely, being near a vehicle that is following too closely, and being in foggy conditions. In addition, the results showed that passenger discomfort can be attributed to a broad range of factors ranging from behaviors of the driver/ vehicle that one is traveling with/in, the behaviors of other surrounding vehicles, the environmental conditions and the vehicle's interior, all of which may differ between different groups of passengers. This research provides important findings and insights into factors that may influence users' acceptance and use of future autonomous vehicles.

Background: Infractions other than collisions are also a crucial factor in autonomous driving since other infractions can result in an accident. Most existing works have been conducted on navigation and collisions; however, fewer studies have been conducted on other infractions such as off-road driving and not obeying road signs. Furthermore, state-of-the-art driving models have not performed dynamic 3D object detection in the imitation learning stage; hence, the performance of such a model is unknown. No research has been conducted to investigate the driving models' computational complexities.

Objective:
The objective of this research is to study the effect of 3D dynamic object detection for autonomous driving and derive an optimized driving model
with superior performance for navigation and safety benchmarks.

Methods:
We propose two driving models. One of them is an imitation learning-based model called Conditional Imitation Learning Dynamic Objects (CILDO), which performs dynamic object detection using image segmentation, depth prediction, and speed prediction. The other proposed model is an optimized model of the base model using an additional traffic light detection branch and deep deterministic policy gradient-based reinforcement learning called Conditional Imitation Learning Dynamic Objects Low Infractions-Reinforcement Learning (CILDOLI-RL).

Results:
An ablation study proves that using image segmentation and depth prediction together to enable three-dimensional object vision improves navigation performance rather than taking decisions entirely from the image. The CILDOLI-RL model presented in this paper achieves the highest
score for the newly introduced No-Other-Infraction benchmark and No-Crash benchmark. It scores a moderate score for the Car Learning to Act
(CARLA) benchmark in both the training town and the testing town, ensuring safe autonomous driving. The base CILDO model achieves the best performance in navigation and moderate scores for safety benchmarks under urban or rural dense traffic environments in both towns. Both proposed models are relatively computationally complex.

Conclusion:
For safety-critical driving, since both navigation performance and safety are crucial factors, it can be concluded that the proposed CILDOLI-RL is
the best model out of the two proposed models. For applications where driving safety is not of much concern, the proposed CILDO is the best
model.

Autonomous vehicles (AVs) are expected to improve road traffic safety and save human lives. It is also expected that some AVs will encounter so-called dilemmatic situations, like choosing between saving two passengers by sacrificing one pedestrian or choosing between saving three pedestrians by sacrificing one passenger. These expectations fuel the extensive debate over the ethics settings of AVs: the way AVs should be programmed to act in dilemmatic situations and who should decide about the nature of this programming in the first place. In the article, the ethics settings problem is analyzed as a trilemma between AVs with personal ethics setting (PES), AVs with mandatory ethics setting (MES) and AVs with no ethics settings (NES). It is argued that both PES and MES, by being programmed to choose one human life over the other, are bound to cause serious moral damage resulting from the violation of several principles central to deontology and utilitarianism. NES is defended as the only plausible solution to this trilemma, that is, as the solution that sufficiently minimizes the number of traffic fatalities without causing any comparable moral damage.

This article presents a detailed exploration of functional safety semiconductor chip design for automotive applications with Functional safety and cyber security capability, focusing on Software-Defined Vehicle (SDV) architectures. The integration of AI/ML accelerators in semiconductor chips is examined to enhance real-time performance, safety compliance, and predictive fault detection. With the increasing complexity of modern automotive systems, meeting (ASIL-B) Automotive Safety Integrity Level B standards is paramount for ensuring reliability and safety in critical systems. We explore the design of a novel semiconductor chip capable of seamlessly integrating AI/ML models, including deep learning, reinforcement learning, and anomaly detection, to predict, identify, and mitigate faults in real-time. The article also emphasizes the importance of fault tolerance mechanisms, including redundant cores and N-Modular Redundancy (NMR), to ensure continued operation during a failure. Furthermore, we discuss how AI-ML accelerators, such as ASICs, TPUs, and FPGAs, can significantly improve processing efficiency, reduce latency, and enhance energy efficiency within the automotive environment. Key considerations, such as cyber security, are also addressed to protect against potential system vulnerabilities while maintaining compliance with ASIL-B safety requirements.

While using fully autonomous vehicles is expected to radically change the way we live our daily lives, it is not yet available in most parts of the world, so we only have sporadic results on passenger reactions. Furthermore, we have very limited insights into how passengers react to an unexpected event during the ride. Previous physiological research has shown that passengers have lower levels of anxiety in the event of a human-driven condition compared to a self-driving condition. The aim of our current study was to investigate these differences in unexpected road events in real-life passenger experiences. All subjects were driven through a closed test track in human-driven and then self-driving mode. During the journey, unforeseen obstacles were encountered on the path (deer and human-shaped dummies appeared). Using physiological measurements (EEG, eye movements, head movements and blinking frequencies) our results suggest that passengers had moderate affective preferences for human-driven conditions. Furthermore, multifractal spectra of eye movements and head movements were wider and blinking frequencies were decreased during unexpected events. Our findings further establish realworld physiological measurements as a source of information in researching the acceptance and usage of self-driving technologies.

In response to the complex demands of autonomous vehicle (AV) navigation in urban environments, this study explores a data-driven, reinforcement learning (RL)-based approach to optimize navigation for AVs, enhancing both efficiency and safety. By integrating data science with RL techniques, particularly Deep Q-Networks (DQN) and Proximal Policy Optimization (PPO), we address challenges such as high traffic density, dynamic road conditions, and complex routing requirements inherent in urban settings. Leveraging real-time traffic data, environmental factors, and city infrastructure, we developed a simulated urban environment to rigorously test DQN and PPO under various real-world scenarios. Our findings reveal that both RL techniques significantly outperform traditional navigation methods. PPO demonstrated superior results with an 89% route completion rate and an 84% improvement in fuel efficiency, compared to 83% completion and 78% fuel efficiency achieved by DQN. These results underscore the potential of RL to revolutionize AV navigation, enabling safer, more reliable, and environmentally sustainable operations in urban logistics. Additionally, the study underscores the critical role of model parameter tuning to optimize RL-driven AV performance and provides actionable insights for real-world deployment of these systems. This research highlights reinforcement learning as a promising avenue for advancing AV technology and supporting the development of smart cities.