Autonomous Vehicles

Cyber-Physical Systems depend on precise, reliable, and real-time sensing to enable advanced monitoring, secure operations, and safety. While classical sensors are characterized by high responsiveness, they are limited to noise and drift faults. Whereas, quantum sensors offer high precision, yet suffer from slow update rates, expensive, and real-time implementation. Existing studies have highlighted the superior accuracy and stability of quantum sensors over classical sensors. However, the integration of classical and quantum sensors is vital to tackle the limitation of individual sensing technologies and to develop more resilient sensing systems. To address this gap, this article proposes a hybrid quantum-classical data fusion framework based on a Kalman filter. The Kalman filter efficiently handles heterogeneous signals with different dynamics and sampling rates, combining the rapid responsiveness of classical sensors with the accuracy of quantum devices. Furthermore, a smart grid system is presented as a proof of concept to demonstrate the practical methodology for the deployment of the proposed framework. Finally, experimental validation is performed to evaluate the performance of the hybrid fusion approach, achieving real-time, robust measurements with an accuracy of up to 94%.

The implementation of a Bluetooth and Camera Based Smart Service Robot using Arduino UNO R3 presents an innovative solution for material transportation in offices and hospitals. Traditional manual transportation of files, laptops, water bottles, and medical appliances requires human effort and consumes time. This proposed robot addresses these issues by automating the carrying process using wireless control and camera monitoring. The robot is controlled through Bluetooth HC-05 using a smartphone application. An ESP32 Camera module is integrated for live video streaming and monitoring. Arduino UNO R3 acts as the main controller, while the L298N motor driver controls four DC gear motors for movement. The robot is designed to carry loads between 2 kg to 2.5 kg efficiently. The total robot weight is approximately 10 kg to 12 kg with a four-wheel chassis for stability and balance. This smart service robot can be effectively used in offices for carrying files, laptops, and bottles, and in hospitals for carrying medicines and appliances.

Multi-sensor fusion has become a cornerstone of 3D perception in autonomous driving systems. However, existing
fusion frameworks commit to a fixed fusion stage—early, mid, or late—regardless of the runtime quality and semantic
consistency of incoming sensor streams. This rigid design fails under real-world adversarial conditions such as camera lens
blur, LiDAR point-cloud dropout in heavy rain, or radar ghosting in urban canyons. In this paper, we present AdapFuse-SD,
an Adaptive Fusion Stage Selection framework driven by a Semantic Disagreement Score (SDS) computed online between
modality-specific feature representations. AdapFuse-SD dynamically routes each inference step to the most appropriate
fusion stage, selecting early fusion when modalities agree at the raw feature level, mid-level fusion when intermediate BEV
features are aligned, and late fusion when per-sensor outputs are more reliable than fused representations. We evaluate
AdapFuse-SD on the nuScenes benchmark and the KITTI-Rain synthetic degradation suite. Our approach achieves a mean
Average Precision (mAP) of 72.8% and a NuScenes Detection Score (NDS) of 75.6%, outperforming the best fixed-stage
baseline by 3.7 mAP points while reducing per-frame inference latency by 8.3% through selective computation. These results
demonstrate that semantic-consistency-aware dynamic fusion is a practical and principled solution to the brittle fixed-stage
fusion paradigm.

This essay is a free foray into the heart of the ontological synthesis of
human-car-society, which manifests itself as the Car-driver, who integrates
the technical being of the car and the bodily and existential being
of the driver with the ideological influence of motorism or 'Cardriving'
as I prefer, with a capital C.

The essay is about the Car-driver hybrid as a fusion of man and
machine in the human-inhabited car, safehouse on wheels, interacting
with society; the subjective possibilities of the human driver when
combined with the technological being of the car; the car as an integral
part of the driver's sensory agency.

The essay revolves around the Car-driver's possibilities and especially
impossibilities when it comes to his interaction with surroundings.
The driver in reinforced isolation, perhaps morally dysfunctional,
deviant but not to blame. The guiding principle is that if there is a specific
behaviour that a Car-driver should exhibit, then it is necessary that
the prerequisites are present. You cannot demand action if the action
cannot be done - 'ought implies can!'. It is these prerequisites
that I am mainly concerned with as they manifest themselves in the
Car-driver.

Haul truck tires represent one of the most significant operational expenditures in open-pit mining, often accounting for 10-20% of total transportation costs (Cunha Rodovalho and de Tomi, 2019; MAXAM Tire, 2025). The surface conditions and temperature of haul roads directly govern the rate of abrasion, heat accumulation within the tire carcass, and the eventual failure mode of off-the-road (OTR) tires. This paper synthesizes peer-reviewed research, industry field studies, and engineering data to examine: (1) the mechanisms by which haul road surface temperature accelerates tire wear and abrasion; (2) the thermodynamic relationship between payload, speed, and internal tire heat expressed through the Tonne Kilometre Per Hour (TKPH) rating; (3) quantified abrasion rates under varying thermal and road surface conditions; and (4) a comprehensive suite of remediation strategies ranging from road design and surface treatment to fleet management, tire selection, and real-time monitoring. The evidence demonstrates that for every 10°F (5.5°C) increase in road surface temperature, tire wear increases by approximately 5% (Li et al., 2017), and that ambient temperatures above 38°C require mandatory TKPH derating of approximately 2% per degree to prevent catastrophic tire failure (Bridgestone OTR, 2023; Magna Tyres, 2025).

Deep learning and AI are used in many ways and are revolutionizing many aspects of our lives. AI in insurance is normally used to determine the premiums for insurance policies, as well as in autonomous cars. This is usually regarded as a principal agent problem. The aim of this paper is to deternine if the business agreements between different AI firms and their partners, such as IBM, Google and others, that compete between themselves in ither fields, may compromise the functioning of AI if the same firm is responsible for, let's say, minimising traffic accidents and maximising profits from policies.

Current legal framework for autonomous vehicles is based on Vienna convention in 1968 that authorises the uses of "self-driving" technology under certain conditions, but requires the presence of a human driver that can take control of the vehicle at any time. In the meantime, the AI technology has become so developed, that the best chess programs are practically unbeatable for human players. Those programs are based on neural networks and training methods that largely depend on the quality of data. The approach to collecting data can be varied. The concept of criminal responsibility is dependent on the awareness of the criminal responsibility, that is usually assumed for humans. In the case of AI, it can not be aware in the human sense, and it evaluates situations in its own terms. It could be licensed as a European product, however, it AI can not really be classified just as a 'product', which means that commercial law should also be improved. The aim of this paper is to assess the suitability of Croatian legal system to be the first to regulate AI for completely autonomous vehicles in the EU, in order to establish a model for this regulation on the EU level. Such a role for Croatia has been proposed, as the Croatian company Rimac Automobili should be the first to produce such a vehicle, and therefore Croatia should be able to provide legal framework for this.

Low-altitude aerial threats, including small unmanned aerial systems (UAS), loitering munitions, and coordinated
drone swarms, increasingly challenge conventional radar-centric air-defence systems operating in contested and
infrastructure-constrained environments. This paper presents an AI-resilient distributed air-defence architecture that
extends probabilistic multi-sensor fusion toward transformer-based multi-modal perception, cooperative multiagent reinforcement learning, autonomous edge-level task allocation, and adversarial-AI-resilient sensing. The
proposed architecture integrates electro-optical, infrared, passive radio-frequency, acoustic, and terrain-context
sensing through a transformer-enabled fusion layer and a cooperative multi-agent coordination layer. The
framework is designed to support resilient counter-UAS surveillance, distributed ISR coordination, and adaptive
sensor management under degraded communication, GNSS denial, node failure, and adversarial perturbation.
Mathematical formulations are developed for multi-modal transformer fusion, adversarial robustness, cooperative
multi-agent policy learning, cue prioritisation, edge-level task allocation, and distributed battlefield sensing. A
Monte Carlo simulation framework is proposed for evaluating detection probability, false-alarm behaviour, latency,
coordination efficiency, node-failure resilience, and adversarial robustness. The architecture provides a scalable
pathway for next-generation distributed air-defence networks capable of supporting resilient counter-UAS
operations within contested electromagnetic environments.

As a critical component of intelligent transportation systems and autonomous driving technologies, traffic sign recognition directly impacts road safety and traffic management efficiency. This study addresses practical challenges in existing CNNbased traffic sign recognition systems, including insufficient accuracy, high computational complexity, and difficulties in small object detection, proposing a comprehensive optimization solution. By introducing lightweight network architectures and channel attention mechanisms, the approach effectively balances model performance with computational efficiency. Incorporating distinctive visual characteristics of traffic signs, targeted data augmentation strategies are designed to significantly enhance the model's adaptability to special scenarios. The adoption of multiscale feature fusion technology improves recognition effectiveness for small-sized targets. Compared to conventional methods, the proposed optimization solution demonstrates notable advantages in recognition accuracy, computational efficiency, and robustness, providing reliable technical support for practical deployment of intelligent transportation systems and holding significant practical value for advancing autonomous driving technologies.

Autonomous vehicles stand at the forefront of technological advancement, promising to alter fundamentally how we navigate, with implications for everything from inter city logistics to daily commuting. The expected benefits of this technology include advancements in safety protocols, streamlining of traffic flow, and superior comfort for those onboard. However, the path to widespread integration is paved with signifi cant barriers that need to be overcome. This article provides a thorough review of the historical development of autonomous driving technologies, beginning with the early vehicles powered by internal combustion engines and extending to the present excite ment over Apple’s Project Titan. Rumors suggest this particular project involves the creation of an electric vehicle fully capable of operating without traditional manual controls, including steering wheels and seats. The article concludes by spotlighting the potential for autonomous vehicles to enable disruptive changes in the transport sector, while also considering the myriad of challenges that must be surmounted for their adoption on a grand scale.

This work investigates vehicular mobility and the main factors that impact Vehicle-to-Vehicle (V2V) connectivity using Basic Safety Message (BSM). MATLAB simulation used for Vehicular mobility and connectivity characterization under specific road traffic conditions. The simulation covers connectivity between traveling vehicles and a selected target vehicle to monitor communication interaction and establish an envelope within which reliable communication and BSM messages can occur. Another objective of this work is to use BSM exchanges to indicate the level of connectivity used to estimate traffic density, thus enabling congestion prediction. The obtained data contain information describing many vehicles, distance, connectivity time, and traffic density. The simulation results indicate an increase in the number of connected vehicles (connectivity level) as a function of both traffic density and communication range. Extending communication over fixed duration showed increased connectivity levels, allowing more vehicles to interact and exchange BSMs. The rate of change of connectivity per communication range is an indication of the state of traffic. Continuous connectivity proved to be less than general connectivity as vehicles exits through ramps and move from one cluster of vehicles to another. Varying duration per fixed communication range produced evidence of spatial domain change, and cluster variation as threshold values separate vehicles clusters in time and space. This work presented a model to help analyze the impact of vehicular mobility as a function of BSM communication range variation and connectivity duration variation correlated to traffic density.

Simulation of Vehicle-to Vehicle (V2V) communication and connectivity is carried out. The main objective of the carried out V2Vcommunication simulation is to study the effect of route length, number of hops per route, attenuation related parameter and message size on energy consumption of transmitted bits per sent message. Mathematical modeling (using the original radio energy model), and analysis, is carried out to quantify and approximate the effect of attenuation related parameter (α) and Route Length (LR) on energy consumption of transmitted Basic Safety Message (BSM) for both, 256 Bytes and 320 Bytes size. The original energy radio model is expanded to include not only message size, but also the effect of number of hops on energy consumption. The work successfully proved the critical effect of α and number of hops on energy consumption for a fixed BSM size and the effect of α on the transitional characteristics of routes as a function of number of hops. It is clear from the simulation that α has an adverse effect on consumed energy for transmitted BSMs and also has a marked effect on routes, where any value of α above 3 will lead to energy depletion among, other negative effects on communicating devices.

The Advanced Driving Assistance System (ADAS) is designed to improve traffic safety and efficiency. ADAS, as a novel solution, is one of the most significant challenges that traffic management will face in the near future It has far-reaching implications, such as the fact that, although it will increase collision avoidance and enable us to move swiftly toward automotive trains, it will also improve railway safety. It has several applications, including in drones and aviation. Autonomous cars and ADAS employ sensors like lidar, GPS, radar, and stereo cameras, making forward collision warning (FCW) for ADAS popular. Vulnerable Road Users (VRUs) such as motorcycles, motorbikes, and Personal Mobility Systems (PMSs)often only utilize a monocular camera on their smartphones or dashboard cameras for safety and navigation, making it hard to get to these sensors. As a result, this paper presents a monocular camera-based VRU FCWs. Without sensor fusion, FCW velocity, collision-time and distance estimates are noisy. The real-time use of SVM and Faster R-CNN vehicle identification algorithms is limited. This work uses the You Only Look Once version 4 (YoloV4) vehicle recognition technique, which uses a stacked Kalman filter to forecast and stabilize velocity, collision-time and distance on monocular camera platforms.

Robust Lane geometry estimation is a critical component of advanced driver assistance systems (ADAS) and autonomous driving, yet its reliability significantly degrades under adverse weather conditions such as rain, fog, snow, and low illumination. This paper presents a self-supervised learning framework for resilient lane geometry estimation without reliance on extensive pixel-level annotations. The proposed approach leverages temporal consistency, geometric constraints, and cross-view photometric alignment to generate supervisory signals directly from raw monocular video streams. By integrating weather-aware feature normalization and uncertainty-guided refinement modules, the model enhances robustness against visibility degradation and sensor noise. A hybrid encoder-decoder architecture with attention-based spatial aggregation captures both local lane markings and global structural priors of road topology. Additionally, a geometry-consistency loss enforces structural coherence across sequential frames, mitigating spurious predictions caused by occlusions and dynamic artifacts. Extensive evaluations on benchmark driving datasets augmented with synthetic and real-world adverse weather conditions demonstrate superior generalization compared to fully supervised baselines trained on limited labeled data. The proposed method achieves improved lane curvature estimation accuracy and boundary continuity while reducing annotation dependency. These results underscore the potential of self-supervised paradigms in enabling scalable, weather-resilient perception systems for next-generation autonomous vehicles.

Mining operations increasingly require intelligent control strategies capable of balancing productivity with energy efficiency, water stewardship, and operational stability. While Mining 4.0 technologies have enhanced automation and monitoring capabilities, most existing systems remain largely reactive and lack intrinsic self-awareness and sustainability-driven decision logic.

This study proposes an Eco-Cognitive Mining System (ECMS), formulated as a hybrid digital twin–surrogate-based, multi-objective cognitive control framework for mineral processing systems. The framework integrates digital twin–based state perception, surrogate eco-physical modeling, and reinforcement learning to enable adaptive, sustainability-aware decision-making under dynamic operating conditions. Energy consumption, water usage, operational stability, and metallurgical recovery are embedded directly within the cognitive optimization process.

Simulation-based evaluation on a grinding–classification circuit demonstrates that ECMS reduces disturbance recovery time by approximately 45% and lowers peak system deviation by about 14% compared to conventional control. Cognitive learning converges within 50–60 episodes, representing a 45–50% improvement in learning efficiency over conventional reinforcement learning. Sustainability trade-off analysis indicates reductions of approximately 11% in energy intensity and 16% in water consumption while maintaining metallurgical recovery. System self-awareness, quantified using a Self-Awareness Index, increases from approximately 0.72 to 0.90, with statistically significant improvement (p < 0.001). These results confirm that ECMS enables self-aware, adaptive, and sustainability-driven mineral processing, providing a practical pathway toward Mining 5.0 systems.

This paper investigates the potential of visible light communication (VLC) for autonomous railway vehicles (ARVs) engaged in nuclear waste disposal. The research focuses on designing an ARV system equipped with VLC technology to ensure safe, reliable communication in hazardous environments, such as deep geological repositories. Initial testing demonstrated that VLC can effectively maintain communication between ARV components over distances of up to 30 meters, even when operating at limiting angles. In this study, angles of 60° and 75° were tested, corresponding to the angles of the curves of the tested route. With 4-QAM modulation it was possible to measure at a distance of approximately 5.5 meters with an angle of 75°. The system has shown promise in addressing key challenges, such as high radiation levels and confined spaces, where traditional RF-based communication systems may fail. The results indicate that VLC could significantly improve the safety and efficiency of ARV operations in nuclear waste disposal, offering advantages in terms of robustness and reliability. Future work will focus on integrating VLC more deeply into ARV control systems, further testing in real-world environments, and exploring its application nuclear waste management.

Panorama stitching consists on frames being merged to create a 360 ° view. This technique is proposed for its implementation in autonomous vehicles instead of the use of an external 360-degree camera, mostly due to its reduced cost and improved aerodynamics. This strategy requires a fast and robust set of features to be extracted from the images obtained by the cameras located around the inside of the car, in order to effectively compute the panoramic view in real time and avoid hazards on the road. This paper compares and creates discussion of three feature extraction methods (i.e. SIFT, BRISK and SURF) for image feature extraction, in order to decide which one is more suitable for a panorama stitching application in an autonomous car architecture. Experimental validation shows that SURF exhibits an improved performance under a variety of image transformations, and thus appears to be the most suitable of these three methods, given its accuracy when comparing features between both images, while maintaining a low time consumption. Furthermore, a comparison of the results obtained with respect to similar work allows us to increase the reliability of our methodology and the reach of our conclusions.

The Pedestrian Safety Behavior Sensor (PSBS) is a novel multi-sensor, edgecomputing platform developed for real-time detection, behavior analysis, and predictive risk assessment of pedestrian-vehicle interactions in urban environments. The system is housed in a rugged, weather-resistant enclosure for roadside deployment and integrates a processing unit (Raspberry Pi or Jetson Nano), RGB camera, proximity sensors, motion detectors, and environmental sensing modules within a unified sensor fusion architecture. The operational framework begins with multimodal sensing, where the camera captures pedestrian movement and trajectories, while ultrasonic or radar sensors estimate pedestrian-vehicle distance. A passive infrared sensor detects human presence under low-light conditions, and environmental sensors monitor lighting and weather factors. These inputs are processed locally using computer vision and machine learning algorithms for pedestrian detection, multi-object tracking, and feature extraction. Temporal features such as speed, acceleration, and trajectory deviation are analyzed to classify behaviors including safe crossing, risky crossing, sudden entry, and congestion. A predictive risk engine combines behavioral, spatial, and environmental parameters to generate a dynamic risk score based on proximity, time-to-collision, and visibility conditions. High-risk scenarios are identified before accidents occur, enabling proactive safety intervention. When elevated risk is detected, the system activates multimodal alerts through LED indicators, audible alarms, and wireless notifications, all within sub-second latency. Additionally, PSBS supports telemetry transmission to centralized dashboards and intelligent transportation systems for large-scale monitoring and high-risk zone identification.

Autonomous vehicles (AVs) integrate heterogeneous sensing modalities-including cameras, LiDAR, radar, GPS, inertial measurement units (IMUs), CAN bus networks, and Vehicleto-Everything (V2X) communication infrastructure-whose collective vulnerability surface presents an unprecedented cybersecurity challenge. Existing intrusion detection systems address individual modalities in isolation, rendering them ineffective against coordinated multi-sensor attack campaigns in which adversaries simultaneously compromise multiple channels to evade detection while steering the vehicle into dangerous states. This paper presents the Intelligent Multi-Sensor Security Analytics (IMSSA) framework-a novel, unified multi-modal threat detection architecture for autonomous vehicles that integrates hierarchical sensor fusion, deep learning-based temporal anomaly detection, a Dynamic Sensor Trust Engine (DSTE), and a Context-Aware Intelligence Layer (CAIL) to achieve comprehensive, real-time cybersecurity across all primary AV sensing modalities. IMSSA introduces three foundational innovations: (1) a crossmodal consistency tensor formalism that captures higher-order statistical dependencies between sensing channels to provide richer anomaly signals than single-modality detectors; (2) a Bayesian Adversarial Evidence Accumulation mechanism that aggregates temporal anomaly evidence weighted by dynamic sensor trust scores, enabling detection of slowly-developing stealthy attacks; and (3) Adaptive Trust-Weighted Fusion that dynamically reconfigures the sensor fusion graph when trust degradation evidence exceeds configurable thresholds, preventing adversarial data injection from corrupting the system state estimate. We construct a comprehensive multi-modal cybersecurity dataset-AV-SecDataset-comprising 847 hours of synchronized multi-sensor data generated through CARLA Simulator integration, covering seven attack scenario categories including GPS spoofing, LiDAR point cloud injection, radar ghost object attacks, CAN bus replay, V2X message manipulation, adversarial traffic sign attacks, and coordinated multi-sensor campaigns. Experimental evaluation demonstrates that IMSSA achieves a detection accuracy of 97.3%, precision of 96.8%, recall of 97.1%, and F1-score of 0.969 on AV-SecDataset-outperforming the best single-modality baseline by 18.4 percentage points on F1. The full multi-modal pipeline operates at 8.7 ms average inference latency on NVIDIA Jetson AGX Orin hardware, satisfying the realtime requirements of production autonomous driving systems. Code, models, and datasets are released as open-source research artifacts.

Smart cities have emerged in the last decade as a testbed for autonomous-vehicle (AV) fleets, with pilot and commercial deployments now running in cities across North America, China, the Middle East, and Europe. This paper surveys the published literature on AV fleet operations in urban smart-city settings, covering foundations of intelligent transportation systems, routing and control algorithms, machine-learning methods for traffic prediction, simulation platforms, and the smart-city infrastructure that supports these deployments. Representative deployments in Phoenix, San Francisco, Los Angeles, Austin, Beijing, Wuhan, Shanghai, Guangzhou, Shenzhen, Singapore, Dubai, Abu Dhabi, Riyadh, and Tokyo are discussed to illustrate how algorithmic advances have been translated into operational fleets. Three comparison tables summarise worldwide deployments, routing approaches, and smart-city infrastructure features. The review concludes that AV fleets in smart cities are progressing rapidly from controlled pilots to commercial operations, and that continued progress will depend on the maturation of routing algorithms, the expansion of V2X infrastructure, and the harmonisation of regulatory frameworks.

The meteoric rise of Agentic AI methods for cyber-physical systems (CPS) in general and autonomous applications in particular, is by no means a coincidence. The challenges of decision-making in CPS (e.g. in autonomous vehicles and robotics) have eluded efforts to develop broad approaches across application verticals. Agentic AI frameworks offer such an approach that combines the data-centric power of machine-learning, with the domain elaboration and contextual reasoning afforded by LLMs, using well-understood decision-making paradigms to construct powerful agents that realize the system's mission.

Autonomous vehicles (AVs) are increasingly positioned in global transport discourse as a transformative solution to urban mobility challenges-promising safer roads, reduced emissions, and efficient movement. Yet this optimism is almost entirely grounded in assumptions about infrastructure that do not hold in Sub-Saharan African (SSA) cities. This paper critically examines the conditions under which AVs could Automation Without Infrastructure: AVs and Sustainability in Sub-Saharan Africa-Ogwo, P.K. environments-without delivering meaningful sustainability returns. A pragmatic sequencing framework is proposed: strengthening road infrastructure, formalizing and greening informal transport, expanding public and active mobility, and improving energy access as preconditions for any meaningful role for advanced vehicle automation in the region.

This study explores the perception on adoption of Artificial Intelligence (AI) for industrial operations from students at Ahmadu Bello University, Zaria, Nigeria. The study achieved the objectives of finding out the perceived role of AI on industrial operations, investigating the perceived benefits of AI on industrial operations and exploring the respondents' concern about the use of AI for industrial operations from students of Ahmadu Bello University, Zaria. A survey research design was adopted using a structured questionnaire by selecting 401 students through random sampling. The findings reveal that majority of the respondents constituting 60.35% reported that AI has positive role in industrial operations. From the respondents, majority of them constituting 69.2 % reported that AI improved efficiency and productivity and most of them also, constituting 56.1%, reported that dependence on technology is a major concern regarding AI use in industrial operations. It is concluded that adoption of AI facilitates industrial operations. It is recommended that integrating AI literacy into the university curriculum, developing strategic partnerships with industry and advocating for national AI education policy, shall be carried out.

The technological landscape in the fields of Autonomous Surface Vessels (MASS), Unmanned Aerial Vehicles (UAVs), and robotics is undergoing a fundamental transformation characterized by the convergence of artificial intelligence, materials science, and cyber-physical systems. The evolution of modern engineering has reached a critical turning point where the boundaries between biological resilience and mechanical robustness are beginning to dissolve. At the heart of this transformation lies a fundamental challenge: How can we protect the next generation of autonomous assets, such as Maritime Autonomous Surface Ships (MASS), Unmanned Aerial Vehicles (UAVs), or deep-sea robotic systems, against increasingly complex and multi-domain threats? This study, entitled "Integrated System Architecture for AI-Powered Autonomous Platforms," stems from the need to move beyond the 20th-century "passive shell" philosophy. Traditional armor and structural design have long relied on static mass and monolithic materials, which are becoming increasingly inadequate in an era defined by high-speed kinetic threats, electronic warfare, and extreme fuel efficiency demands. We propose a revolutionary framework: "Armor Immunity System." The need for next-generation platforms exhibiting superior strength-to-weight ratios while maintaining high survivability levels has led to the emergence of the "Armor Immunity System" concept. This architecture goes beyond traditional passive shielding, creating a dynamic, forensic, and reconfigurable protective framework. Like a biological organism, an autonomous platform must be capable of sensing structural trauma, identifying the "DNA" of its components for forensic integrity, and adapting its defensive posture in real time. By combining the latest advancements in AI-powered material computing, high-entropy alloys (HEA), and Level 5 Digital Twin synchronization, this research presents a technical blueprint for a self-aware, self-defending architecture. This study provides a comprehensive technical analysis and synthesis of the components required to realize such a system, integrating AI-powered material discovery, high-accuracy digital twins, additive manufacturing of functionally graded materials, and advanced cloaking metasurfaces. It serves not only as a technical synthesis for maritime and aviation professionals but also as a strategic manifesto for the future of cyber-physical survivability. It bridges the gap between theoretical materials science and the operational realities of 21st-century autonomous warfare and logistics.

Rapid urban growth and rising traffic complexity require Smart City solutions that move beyond passive CCTV toward intelligent, real-time traffic management. This study examines how computer vision-based analytics contribute to road safety when integrated into an Intelligent Transportation System (ITS). A quantitative quasiexperimental design was applied across multiple intersections using a 12-month beforeafter window. Data were collected from video analytics (vehicle and pedestrian detection, tracking, violations, road conditions), adaptive signal logs, crash and injury records, near-miss indicators, and contextual variables such as weather and traffic volume. Analysis combined perception validation (mAP, tracking accuracy), time-series operational assessment, and Difference-inDifferences modeling to estimate safety impacts. Results show high perception reliability (mAP > 0.85) and significant operational improvements, including a 33% reduction in waiting time and 35% shorter queues. More importantly, red-light violations decreased by 39%, near-miss events by 45%, crash frequency by 42%, and severity index by 37%. The findings indicate a causal pathway from vision-based perception to adaptive control and enforcement, leading to measurable safety gains. The study concludes that computer vision serves as a safety governance instrument within Smart City ITS when detection outputs are tightly coupled with intervention mechanisms.

The development of an unmanned spacecraft to transport people and equipment out of the Earth has been successfully completed with recent missions to the International Space Station, confirming that the technology is almost ready for farther destinations.
Plans for starting regular landings on the Moon will take 5 to 10 years from present time. After essential devices and communications systems are deployed, the arrival of humans is a matter of months. In the meantime, autonomous equipment with auto-assembly capabilities could start the first terrain-leveling activities for pad building and allocation of containers and objects. The main purpose will be to develop the first settlements for explorers, including communications technicians, assembly crew, geologists, engineers, machine operators, and mechanical/electronic support personnel.
One of the main activities that will drive the presence of people is to start mining operations in the shortest time possible, as this will become the source of materials for infrastructure construction and industrial manufacturing.
Although building settlements is a primary need, the process may take some time that could be balanced with temporary inflatable hangars for living, for mechanical shops and storage spaces. Once the settlement has been completed, mining operations will become the main activity.

The growing integration of artificial intelligence (AI) into wireless communication systems is driving a shift toward semantic communication, an emerging paradigm that prioritizes the exchange of meaning over raw data. However, semantic communication systems face major challenges when deployed across diverse and unseen domains due to variations in language, context, and channel conditions. This survey provides a comprehensive overview of Domain Generalization (DG) as a key enabler for improving the robustness and adaptability of AI-enabled semantic communication. We explore the types of domain shifts and review the latest DG techniques applicable to semantic communication. Additionally, the paper discusses architectural considerations and real world applications across varied wireless scenarios. Unlike prior works, this survey brings together DG strategies specifically within the context of semantic communication, identifying open challenges and future research directions such as scalable adaptation, resource efficient deployment, and resilience in dynamic environments. It aims to serve as a timely resource for researchers and practitioners working to develop reliable, generalizable communication systems for next generation networks.

Technological crises refer to critical events-system failures and collapses, cybersecurity and data breaches, software meltdowns, critical infrastructure disruptions, etc.,-in which the failure, malfunction, or disruption of technological systems creates 0 0 1 E significant risk to safety, operations, or the environment. These crises typically arise unexpectedly and require rapid, well informed 0 0 1 E decision making under conditions of uncertainty, time pressure, and missing or incomplete information, to prevent escalation. Artificial intelligence (AI) systems have emerged as powerful tools capable of augmenting human judgment in these contexts. This article examines the 0 0 1 E role of AI in supporting human decision making during technological crises, analyzing its capabilities, limitations, and implications for organizational resilience. It emphasized that AI is most effective when used as a collaborative partner rather than a replacement for human 0 0 1 E expertise, and it outlines design principles for trustworthy, human centered AI systems.

This paper investigates the connectivity of vehicles to road-side infrastructure (V2I) using a mmWave system at 60 GHz. Two highways in the UK were selected as virtual test drive routes. Detailed 3D geographic data of the city are incorporated into a 3D mmWave ray-tracing tool for the purpose of generating realistic channel matrices for use in an IEEE 802.11ad physical layer (PHY) simulator. The spatial and temporal multipath components of the radio channel between a road side unit (RSU) and the vehicles are modelled. Up to half million point-to-point ray tracing simulations were performed. Important parameters such as delay and angular spread for channel modelling have been extracted from the results. Statistical analysis for each route is presented. Spatial and polarimetric convolution between the multipath ray data and the appropriate beam pattern is performed in order to generate throughput predictions using a link level abstraction simulator. The throughput performance along the routes is presented for the Single Carrier (SC) transmission scheme of IEEE 802.11ad. A comparison between isotropic antennas and phased array beamforming is presented. It was found that three 60 GHz RSUs provided full coverage for a 2 km highway route. The simulation results provide a useful insight on the RSU deployment requirements.

Hydrogen vehicles are facing significant challenges, including safety risks in the event of leakage, high costs associated with physical testing, and fuel cell inefficiency under dynamic conditions; to address these concerns, we propose VR-H2Safe, a cyber-physical platform that combines high-fidelity simulation, physics-informed reinforcement learning (RL), and secure vehicle-to-everything (V2X) communication. Using ANSYS Fluent Large Eddy Simulation (LES) to train real-time predictive models, the platform's digital twin simulates hydrogen leakage with 92% spatio-temporal correlation compared to high-fidelity LES-CFD benchmarks at key monitoring points. Furthermore, we introduce the H2-LEAK protocol, which extends the SAE J2735 standard with a 128-bit risk vector and blockchain authentication, which provides hazard warnings in less than 2 seconds with 95% spoofing resistance. In addition, our physics-informed AI system optimizes the fuel cell performance, achieving a 10% increase in energy efficiency and a 58% reduction in thermal violations during hardware verification. The experimental findings of urban, tunnel, and road conditions prove that VR-H2Safe accelerates the process of safety verification and reduces the engineering validation phase from 14 to 10 weeks.

Latin American countries experienced a significant reduction in income inequality at the turn of the 21st century. From the early 2000s to around 2012, the average Gini coefficient fell from 0.514 to 0.476. The period of falling inequality coincided with leftist presidential candidates achieving electoral victories across the region: by 2009, eleven of the seventeen countries had a leftist president -the so-called Pink Tide. We investigate whether there was a "leftist premium" on the decline in inequality and, if there was one, through which mechanisms. Using a range of econometric models, inequality measurements, and samples, we find evidence that leftist governments lowered income inequality faster than non-leftist regimes, increasing the income share captured by the first seven deciles at the expense of the top ten percent. Our analysis suggests that this reduction was achieved by increasing social pensions, minimum wages, and tax revenue.

Capability projection produces deployment failures whose proximate symptom is the decorative captain-a human assigned operational responsibility for an automated pipeline whose pace, scope, opacity, or speed of state change has already exceeded what the human's authority can connect to. Paper I named the cognitive error and prescribed pipeline analysis plus the captain principle. Paper II refined the captain principle through three historical cases, identified three structural refinements (visibility, credibility, transparency), and sketched four broader failure categories that the trilogy's third volume develops. This paper completes the framework by working through the four broader categories with their historical anchors and contemporary AI parallels-hidden automation, handoff and skill atrophy, speed asymmetry, trust miscalibration-by examining counterexamples in domains that paid for their captaincy rulebooks (post-TMI nuclear, modern commercial aviation, surgical robotics, pharmaceutical adverse-event reporting), and by addressing the central operational question Papers I and II named but did not resolve: who is the captain in a contemporary AI deployment, and what does the captaincy look like across the deployment types currently bundled under the term artificial intelligence. The paper develops the captaincy architectures for predictive, generative, and agentic AI, with concrete pre-authorization mechanisms (bounded action spaces, escrowed execution, reversible-first design) for the agentic case; specifies the four minimum conditions under which distributed captaincy still operates as captaincy; splits trust calibration into humanside training and system-side interface design; distinguishes raw disclosure from actionable interpretability; addresses the incentive incompatibility that makes voluntary adoption rare; and provides a captaincy audit checklist as the framework's diagnostic surface. The paper closes the trilogy by synthesizing its diagnostic, refinement, and operational layers into a single deployment discipline applicable across predictive, generative, and agentic AI systems, naming the boundary condition beyond which structured captaincy may not scale, and locating the open work the framework does not complete. The substrate has been written by older industries. The choice the AI era faces is whether to inherit it, repeat it, or pretend the lessons do not apply.

The global maritime sector is undergoing an unprecedented period of transformation, characterized by the convergence of deep decarbonization imperatives and the rapid evolution of autonomous navigation technologies. This paradigm shift is most evident in the revival and modernization of Wind-Assisted Propulsion Systems (WAPS), evolving from traditional fabric sails to advanced, computer-controlled aerodynamic engines. However, the modern revival of wind power differs fundamentally from its historical predecessor. Contemporary WAPS utilize advanced aerodynamic principles such as the Magnus effect, boundary layer suction, and high aspect ratio rigid wings, all managed by complex automated control systems. This shift marks a transition from "sailing" as a manual tool to "wind-assisted" as a precision engineering discipline. By early 2026, the global order book for commercial vessels had reached 191 million compensated gross tonnage (CGT), the highest level in 17 years, with a significant proportion of newly built vessels integrating alternative propulsion systems or energy-saving devices. Between 2021 and 2025, the industry focused on proving the feasibility of modernizing existing vessels, enabling approximately 75% of the current WAPS fleet to consist of modernized ships. By 2027, the focus has shifted to "primary wind" designs, where the hull, machinery, and control architecture are optimized from the ground up to maximize wind energy harvesting. This evolutionary phase is characterized by the integration of WAPS into hybrid-electric power systems, with wind acting as the primary propulsion force and electric motors becoming the secondary driver providing additional power. This has resulted in a more predictable and efficient operational profile for WAPS. In this context, WAPS is now seen not merely as an increase in environmental efficiency but as a fundamental pillar of maritime fuel transition performance, providing a predictable and inexhaustible energy source that offsets the high costs of near-zero emission fuels such as green ammonia and methanol. The integration of advanced aerodynamics, particularly at levels 3 and 4, within the framework of Maritime Autonomous Surface Vessels (MASS) represents the frontier of modern maritime architecture. The decision to move towards autonomous models is increasingly influenced by the complex physical and environmental impacts of highly efficient wind technologies such as Flettner rotors and suction sails on human crews. Issues related to bridge line-of-sight obstructions, acoustic disturbances, and low-frequency vibrations necessitate a transition to "Digital Captain" architectures where AI-powered sensing layers replace human sensory limitations.

The article describes work in Finland in forming international research group/network on activities relating to autonomous and remotely operated vessels, which concentrates in tackling marine business related obstacles and solutions especially for risk management, marine insurance and charter parties legal issues. Comparative approach to these issues connects national and international experts to network and future research alliance in MET. The purpose of the activities is to create a network of experts which will seek funding for future work to enhance Nordic model of risk management for autonomous vessels. The Nordic Marine insurance Plan already has a relevant market share in marine insurance worldwide, but the specific research area is promoting applications for research activities which lead to the Nordic solution to be the market leader in marine insurance for autonomous and remotely operated vessels worldwide. The article describes the benefits of this approach in relation to other jurisdictions. This approach will also concentrate on facilitating the maritime transport by renewing the charter party clauses for the benefit of those operators who use the Nordic Marine insurance plan for the basis of their risk management strategy and possibilities to use cost effective Nordic Offshore and Maritime arbitration (NOMA). The article is part of promoting research activities in forming research alliance and networking of experts internationally for international maritime transportation.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Intelligence, smart sensors are very much required to link with naval systems in current era to take autonomous decision making and enhancing the capabilities of existing systems. Artificial Intelligence (AI) based algorithms and deepcatch models are making predominant changes in the existing sensing technologies with much more precision in a real-time environment which will revolutionize Navy. But there are lot of challenges and features during the incorporation of AI in the performance upgradation of naval platforms. The latest improvement in the AI proactive real-time intelligence algorithms exceeding the human intelligence capability with the help of Merlin Deep Learning (DL) methods/models to extract and analyse the sensor parameters reading from the naval systems. Nowadays, Internet of Things (IoT) sensors installed with Reinforcement Learning (RL) algorithms are used to detect and process the object information in the naval systems. This paper discuss the key factors involved in the incorporation and upgradation of current sensor systems with their issues, challenges, benefits and features. The ideas and concepts presented in this paper are merely intended to provoke and stimulate thinking about human-machine interaction teams that would function on the AI-enabled battlefield of the future.

We present a reproducible AI-driven framework for autonomous highway control integrating V2X communication and IoT sensing. A hybrid model combining LSTM-based traffic prediction and reinforcement learning (RL) for control is proposed. We provide a benchmark against state-of-the-art (SOTA) baselines (DQN and multi-agent RL), an ablation study, and statistical validation over multiple random seeds. Evaluated in SUMO with European-like highway scenarios, the framework achieves +29-33% throughput improvement,-32-38% collision probability reduction, and-12-16% CO2 emissions. Results are statistically significant at 95% confidence. Code and configuration details are specified for reproducibility.

Safety-shielded neural planning has emerged as a promising approach for enabling reliable and human-centric decision-making in urban autonomous driving. This method integrates data-driven neural networks with formal safety mechanisms to ensure that generated driving actions remain within predefined safety constraints. In complex urban environments characterized by dense traffic, pedestrians, and unpredictable behaviors, purely learning-based systems may produce unsafe or non-compliant actions. To address this, a safety shield acts as a supervisory layer that monitors and corrects the outputs of neural planners in real time. The proposed framework emphasizes human-centric driving by incorporating social compliance, comfort, and interpretability into the planning process. It leverages contextual perception, intent prediction, and rule-based safety verification to produce trajectories that align with human driving norms while maintaining strict safety guarantees. Additionally, the approach supports adaptability to diverse traffic conditions and improves robustness against uncertainty and edge cases. Overall, safetyshielded neural planning provides a balanced solution that combines the flexibility of machine learning with the reliability of formal safety systems, making it a critical component for deploying trustworthy autonomous vehicles in real-world urban settings.

This is highly recommended to manage the traffic in urban transportation as consideration of the busyness of the roads in business hours and less awareness as well as less obedient people for traffic rules in public. In this scenario faster vehicle movement and the rout selection are the challenges. As the solution of this problem may be address by the advanced graph based algorithm as well as machine learning techniques. This paper is focuses on the comparative study of 2 major techniques for routing as like Imitation Learning (termed as IL), Anytime D* (termed as AD*) algorithm. This study concentrate on the comparison of performance, computational complexity, adaptability and feasibility for deployment. The various results manifest that AD* is faster and accurate in between source to the destination for use in robotic navigation in dynamic environments. Where as the IL has ability to train the model based on the advanced planning techniques or the behaviour of expert drivers behave. The new approach requires to be designed which may more effective for finding the shortest path with learning driven techniques for the intelligent road transportation.

Drawing perspectives from science and technology studies, philosophy of science, and literature from ethics and social justice, this paper examines the promises and challenges in the development of self-driving vehicle (SDV) technology. We start with the premise that the combination of different computing technologies embedded in SDVs is a powerful tool for efficiency in communications, information gathering, processing, and storage. However, by focusing on efficiency, SDVs provide a new mode of industrialized transportation whose users can only choose between transportation services, but have little or no say about the broader social implications of the technology. We argue that perspectives from social justice and ethics show that SDVs have implications beyond transportation, with profound consequences for users and societies. In particular, values such as privacy, security, and responsibility may be changed for good or bad, in both the short and long-term. The examination of these changes, while the technology is still under foundational development, is as urgent as it is needed.

Traffic congestion is a burgeoning issue within cities, prompted by rising populations, aging infrastructure, and inefficient traffic management systems. Current AI-based models for traffic control are mainly based on deep learning or reinforcement learning, which, as powerful as they are in prediction, have high computational complexity and poor interpretability. In contrast, rule-based symbolic AI is explainable but poor at real-time adaptation. To bridge these gaps, we introduce the NSFE, a new hybrid AI framework that combines transformer-based deep learning, rule-based GNNs and symbolic logic, and reinforcement learning to boost realtime traffic optimization decision-making. The novelty of NSFE is its attention-based fusion process that balances dynamic neural perception with symbolic reasoning while ensuring interpretability and accuracy. Deep insight module learns fast traffic patterns, while the symbolic reasoning module maintains logical constraints as well as regulation adherence. Ongoing reinforcement learning loop optimizes traffic control maneuvers at all times and adapts to changing situations. Edge computing integration maximizes computational effectiveness in order to accomplish real-time handling with zero delay. Experimental evidence shows that NSFE surpasses conventional DRL and rule-based hybrid models, realizing up to 15% better accuracy, 30% lower computational overhead, and improved resistance to anomalies. Through the fusion of high-precision forecasting with structured reasoning, NSFE delivers a scalable, interpretable, and efficient AI-based solution to urban traffic control. Apart from traffic control, its structure has wider applications in autonomous systems, finance, robotics, and IoT, hence being a revolutionary methodology to real time decision making.

The maritime industry has evolved beyond being merely a physical transportation sector; it has become the forefront of the 21st century’s most advanced computational challenges. This research paper presents a doctoral-level synthesis of two rapidly converging fields: the maturation of Quantum Operating Systems (QOS) and the legal-technical implementation of Maritime Autonomous Surface Ships (MASS) by the year 2032. The maritime industry is undergoing a structural transformation that goes beyond mere digitalization and is entering an era defined by the convergence of quantum computing, high-precision synthetic intelligence, and autonomous physical architectures, We are approaching a transition from “Digitalization” to “Quantumization,” where the complexities of ocean physics, global logistics, and cybersecurity are addressed through the unique logic of quantum bits and hybrid orchestration. From this perspective, while the limits of classical computing—defined by the stagnation of Moore’s Law—threaten the scalability of complex maritime operations, the emergence of the Quantum Operating System (QOS) will serve as an interface to facilitate the necessary technological leap. This study will serve as a strategic template for engineers, regulators, and stakeholders to design an ecosystem that is not only autonomous but also fundamentally “quantum-secure” and environmentally sustainable. In a sense, this study analyzes the trajectory of maritime technology toward 2032, focusing on the integration of quantum-classical-intelligent hybrids in navigational decision-making, robotic swarms, and digital twin infrastructures, thereby establishing a multidisciplinary framework for the implementation of the Quantum-Maritime Operating System (QM-OS). Furthermore, this study presents a methodological roadmap for “Hybrid Compilation Coordination”—essential for managing heterogeneous resources, including QPUs, GPUs, and CPUs, in real time—by integrating a Quantum-Maritime Operating System (QM-OS) modeled after pioneering frameworks such as “Origin Pilot.”

The global maritime sector is currently navigating a structural metamorphosis, transitioning from localized automation to the threshold of Truly Autonomous Surface Ships (MASS). This strategic roadmap toward 2032 explores the necessity of a fundamental reorganization in computational logic, material engineering, and robotic autonomy. Central to this transformation is the emergence of the Quantum-Maritime Operating System (QM-OS), an integration layer designed to manage the "missing middle" of the quantum computing stack. By orchestrating hybrid workflows between classical CPUs/GPUs and Quantum Processing Units (QPUs), the QM-OS bridges the gap between heterogeneous hardware and real-time operational demands. This study provides a multidisciplinary analysis of the application areas for quantum-enhanced maritime systems, focusing on construction, repair, AI-robotics integration, and digital twin development. The shift toward 2032 is not merely an incremental upgrade but a "neuro-robotic convergence" where the limits of human cognition are expanded through Proto-AGI and high-intelligence ecosystems. This study details the technical implementation of such systems, from Project Poseidon’s use of Quantum Approximate Optimization Algorithms (QAOA) for generative hull design to the A.G.E.I.S. framework’s deployment of robotic agents for high-vibration engine environments. By dismantling traditional operational silos, this roadmap establishes the foundation for a "system-of-systems" approach, ensuring that future maritime trade is built upon a sovereign and quantum-safe infrastructure.

Traffic safety in Boyolali Regency is at a critical juncture: on the one hand, the system has shown progress in reducing fatalities; on the other, the number of accidents has actually increased significantly. Data from 2024-2025 recorded an 18.5% increase in the number of victims, from 2,039 to 2,416, while fatalities decreased by 13.2%, from 129 to 112. This phenomenon marks a fundamental shift in the nature of traffic risks—from severity to frequency.