ABSTRACT. Accommodating pedestrians crossing midblock has been shown to have harmful environmental consequences because of increased fuel consumption and CO2 emissions. Somewhat surprisingly, no studies were devoted to mitigating the environmental impact of midblock crossing. Our main contribution is to propose schemes that mitigate the increased fuel consumption and CO2 emissions due to pedestrian midblock crossing by leveraging information about the location and expected duration of the crossing. This information is shared, in a timely manner, with approaching cars. We evaluated the impact of car decisions on fuel consumption and emissions by exploring potential trajectories that cars may take as a result of messages received. Our extensive simulations showed that timely dissemination of pedestrian crossing information to approaching vehicles can reduce fuel consumption and emissions by up to 16.7%.

ABSTRACT. The complexity of controller design for achieving high agility and speed in maneuvers arises from the non-convex nature of the objective function, reliance on noisy parameter evaluations during tuning, and the need for dynamically changing parameters across different trajectory segments. Building upon these challenges, this paper presents a comprehensive framework for the controller design process, addressing the complexities inherent in achieving high agility and speed in maneuvering quadcopters. The proposed design philosophy is built upon a range of research conducted over the last 15 years, consolidating insights and advancements from diverse studies in the field. It begins by discussing early demonstrations of vision-based autonomous flight and progresses to recent investigations into human vs. algorithmic piloting abilities. Key challenges such as perception-to-action dependence, domain shifts, and sensor limitations are highlighted, along with proposed solutions including hybrid simulation-real world approaches. In addition to the comprehensive control system design philosophy, it also evaluates potential drone platforms for research and development purposes. Through this thorough examination, the paper provides a foundation for further studies in agile controller design for quadcopters, highlighting the importance of robust methods and innovative approaches in advancing drone autonomy. This framework provides a basis for further exploration and development in the field, with potential applications across various controller design methodologies.

ABSTRACT. Estimating fuel consumption is an important aspect of promoting smart mobility. Fuel consumption can be calculated based on energy demand models. Instantaneous energy demand models require data about car speed at each second. This provides an accurate estimate of the instantaneous fuel consumption. On the other hand, energy demand can be estimated using the average speed for the whole trip. This approach is less accurate than the instantaneous approach, but it is faster and more convenient to use. In support of smart mobility, this paper takes a close look at the difference noticed while estimating the energy demand using instantaneous and average speeds in on-ramp scenarios. Our theoretical results show that the acceleration rate is an important factor that affects the accuracy of the average energy demand model. This was confirmed empirically by looking at a number of recent car models. The empirical results closely match and support our theoretical findings.

ABSTRACT. Electricity demand in Kingdom of Saudi Arabia is indicating increasing and decreasing trend. The reason behind this erratic growth is government policies in the country. Basically, about forty percent of Saudi population is consisting on non-Saudi nationals having residency rights in the country. This major chunk of population is directly linked with government policies leading to reduced population in various years. Similarly, Grand Domestic Product (GDP) of the country is affected and it also affects the energy demand in the country. Though, for a long period of time overall demand shows increasing trend but it limits its increasing capability and hinders economic growth. Long-range energy alternative Planning System (LEAP) is used for demand estimation model in this analysis. Three scenarios are designed to find the clear picture of demand. These are Business as Usual Scenario (BAUS), High Growth Scenario (HGS) and Demand Management and Energy Efficiency based High Growth Scenario (DMEEHGS). Analysis indicates that energy demand will be 336.2 TWh (Tera Watt hours) in 2040 in BAUS whereas this demand will be 631.4 TWh in HGS. By applying demand management and energy efficiency on HGS demand will be 511.4 TWh. In terms of cost about 21.6 Billion Saudi Riyals can be saved by applying demand management and energy efficiency on High Growth Scenario. This clearly indicates that investment on Energy Efficiency will be more beneficial than the investment on catering high peak demand.

ABSTRACT. —This paper presents an optimal control scheme for connected and autonomous vehicles (CAVs) through double adjacent, multilane signal-free crossings. The objective is to reduce energy usage and enhance traffic flow. The framework consists of two planning levels. During high-level planning, each connected autonomous vehicle (CAV) calculates the best arrival time and lane selection at every junction to improve traffic flow. During the low level planning stage, a control issue focused on optimizing energy use is defined. By resolving this problem, the optimal acceleration or deceleration for every Connected and Automated Vehicle (CAV) to traverse the intersections within the specified time determined by the upper-level planning is ascertained. This is subsequently contrasted with the baseline scenario to demonstrate the efficacy of the control method.

ABSTRACT. Waste, as a primary cause of visual pollution, not only impacts public health but also has significant economic implications, particularly in tourism. Visual pollution from waste or trash encompasses various types that require classification. Cognitive cities are beginning to develop automatic systems to classify these types, but the task is challenging due to the similarity among different types of waste and the common features of most elements. To address this issue, we propose an innovative approach using the YOLOv8 object detection model to detect 16 different types of trash. The proposed approach is compared to the traditional YOLOv8 to evaluate its performance. The results demonstrate the potential of the modified YOLOv8 approach, particularly when applied to larger image sizes achieving a notable improvement in F1-score, underscoring the viability of the proposed approach.

ABSTRACT. Demand forecasting is one of the most critical steps in planning. The success of the operations and customers’ services highly depend on reliable forecasts; however, forecasting is not easy, especially when the time series is highly volatile. Intermittent and lumpy demands are even more challenging to forecast due to zero values in the observations. In last-mile logistics, when the product variety is high, both fluctuation and intermittency of the demand are faced. Forecasting such demand becomes extremely crucial. This paper uses machine learning techniques to forecast demand for 160 different products for 120 different clusters in Riyadh metropolitan area, Saudi Arabia. Our models forecast daily demand for each product in each region, predicting more than 4000 entries. Our results showed that the machine learning models perform much better than traditional methods like Croston’s method.

ABSTRACT. Solar PV panels are widely used solar energy harnessing technology globally. However, after the end-of-life (EoL), a tremendous amount of waste will be generated in many countries, forcing policymakers to give special attention to managing the waste stream. Recovering the stream and subsequent process falls under the category of reverse logistics (RL), which plans for waste allocation to dedicated recycling facilities. Saudi Arabia is heavily investing in installing large-scale solar PV plants, and as of 2022, 390 MW of solar PV generation capacity has been installed. This paper aims to develop an RL network design for future Saudi Arabian solar PV waste that will be generated in the year 2043 using a mixed-integer programming (MIP) optimization model to minimize the total cost of the network. A 24 (solar plant) × 24 (recycling center) distance matrix has been developed as model input. In addition, the waste generation amount from individual solar PV plants has been estimated using the Weibull-distribution-based forecasting technique. Results of the study showed that the total optimized cost of the network will be around Saudi Riyals (SAR) 1.8 billion, including SAR 7.1 million in transportation costs, SAR 910 million for the recycling process, and a capital investment cost of SAR 922 million. It is also estimated that from the transportation process, total CO2 emission will be around 12 million kg. Based on the model, it is also found that the required number of recycling centers will be needed for Eastern Province (4 units), Al-Jouf Province (3 units), Makkah Province (6 units), Medina Province (3 units), Riyad Province (5 units), Al Qassim Province (2 units) and Northern Border Province (1 unit), each with a capacity of 32000tons/year. This paper will also help policymakers and waste management authorities make preliminary decisions on financial and physical resource allocation.

ABSTRACT. Drones have become ubiquitous in various industries due to their versatility and efficiency in performing a wide range of tasks, from agricultural operations to search and rescue missions. This paper explores the use of drones, particularly cargo drones, in revolutionizing logistics and transportation systems. Medium-range cargo drones offer the potential to transform freight transportation by providing independence from traditional infrastructure and minimizing environmental impact. However, their integration into existing logistical operations faces several challenges, including regulatory hurdles, technological limitations, and public perception issues. Despite these challenges, cargo drones offer numerous benefits like cost savings, additional mobility, flexibility, and environmental friendliness. Drones can prove to be a viable form of cargo transportation given that the regulatory challenges are addressed and can be efficiently integrated into the existing logistic operations. It would ultimately result in an efficient last-mile delivery option and will revolutionize the logistics industry.

ABSTRACT. The surge in popularity of on-demand transportation services, fuelled by advancements in technology and changing urban mobility patterns, has significantly reshaped urban transportation dynamics. However, amidst this transformation, a critical gap persists in understanding the intricate interplay between on-demand transportation systems and the broader urban transportation ecosystem. This gap presents a pressing problem statement, as the rapid growth of on-demand services challenges traditional public transportation while raising questions about sustainability and energy efficiency. The aim of this research is to address this gap through an explorative literature review. The methodology involves systematically reviewing existing literature on on-demand transportation systems, focusing on key themes such as operational efficiency, energy transition, and policy implications. By synthesizing and analysing this body of literature, the research aims to uncover insights into the current state of on-demand transportation, identify challenges and opportunities, and propose avenues for further research. Furthermore, this study seeks to develop operational and theoretical framework to support policy formulation and implementation in urban transportation planning. By integrating findings from existing case studies with insights gained from the literature review and practical considerations, the frameworks will offer guidance for policymakers and urban planners aiming to promote sustainable, energy-efficient on-demand transportation systems. Ultimately, the research endeavours to contribute to the development of evidence-based policies and practices that foster the sustainable development of urban transportation networks in response to evolving mobility trends. The study reveals key insights into the impact of on-demand transportation services on urban mobility, highlighting challenges such as operational efficiency, energy transition, and policy implications, while proposing operational and theoretical frameworks to guide sustainable policy formulation and implementation in urban transportation planning.

ABSTRACT. Magnetic levitation systems, also known as Maglev, are electromechanical devices that employ electromagnetism to suspend ferromagnetic materials. These systems have garnered increasing attention due to their ability to eliminate energy loss caused by friction. Primarily aimed at reducing friction, Maglev systems find extensive applications in various engineering domains like magnetic bearings, precise positioning platforms, aerospace shuttles, and rapid Maglev trains. This research focuses on the real-time identification of a Maglev model using a neural network. Real Experimental data from the 33-210 Maglev system is used to identify the system with shallow and deep neural networks. The shallow neural network produces a regression of 0.99986 and a performance of 9.2e-05 mean square error (MSE). Comparatively, the deep neural network exhibits superior results with a higher regression of 0.99987 and improved performance at 8.456e-05 (MSE).

ABSTRACT. This paper presents a comparative analysis between a two-echelon supply chains (SC) with variable production rates in traditional (TRD) system and a constant production rate in Vendor-Managed Inventory (VMI) system. The first echelon comprises a vendor (supplier), while the second echelon consists of a single buyer. The VMI approach allows the supplier to initiate purchase requisitions rely on demand data given by the client. Our study utilizes the VMI model of Razmi et al (2010) and the findings show that initially, the total cost for VMI system (TVMI) is lower than the total cost for TRD system (TTRD) when demand (D) decreases, However, as D increases, TTRD becomes lower than TVMI. This reflects the cost dynamics with D changes. Additionally, sensitivity analysis indicates that as parameters increase, both order quantity (Q), production rate (P), reordering point (r), and total costs increase, with VMI consistently outperforming TRD in cost efficiency, leading to savings. Overall, they indicate that VMI reduces inventory costs and provides greater cost-efficiency than TRD systems. Reduced buyer ordering expenses further highlight VMI’s benefit, promoting its broader adoption in supply chain management.

ABSTRACT. The exponential growth of e-commerce has made it increasingly important for home delivery services to optimize the delivery experience for ultimate customer satisfaction. A key aspect of this optimization journey is negotiating and assigning delivery time windows between carriers and customers, particularly in the crucial last-mile delivery phase. Broader delivery windows allow carriers to accommodate flexibility in their operations, while customers tend to prefer narrower time frames to minimize waiting periods. In this paper, we address the challenge of Dynamic Vehicle Routing with Time Window Assignment (DVRPTW), which is a fundamental problem in modern logistics management. Our approach involves introducing a dynamic strategy for assigning time windows to customers, starting with providing ample time windows for established customers and then fine-tuning them in response to new orders and existing route constraints. The paper provides interesting insights into the description of this novel method and highlights its contribution to addressing the above issue, while also suggesting directions for future research.

ABSTRACT. The government of Botswana has been and is encouraging the use of technology to exploit opportunities offered by the Fourth Industrial Revolution (4IR) to the different economies of the country. Agriculture in Botswana is one economy that still faces many challenges such as communication, electricity, water, animal theft, farm Management, animal health and diseases, vaccination records, marketing and many others. 70-80% of the rural population in Botswana depend on farming for their living sustenance. Smart Innovative technology developments have the potential to address some of these Agri-Business challenges in Botswana and make farm management more efficient. This paper presents some block diagrams of innovative Internet of Things systems that can be developed to address some of these challenges. These include Water and Diesel flow rate and Level Monitoring Systems; Inventory Systems; Soil Management Systems, Environmental Management Systems; Animal Health and Tracking Management Systems. These technologies have the potential to be developed and used in agri-business.

ABSTRACT. This research introduces a queuing framework, denoted as M1/M2/c, to analyze the dynamics of electric vehicle (EV) charging stations. The framework considers non-uniform Poisson arrival rates (M1), influenced by road traffic conditions, and exponential service durations (M2), affected by factors such as battery size and charging habits. The variable "c" represents the number of chargers at the station. Using this framework, the study evaluates the charging demand and capacity of a lone EV charging station. Results indicate that the station's capacity to simultaneously charge EVs is limited, despite varying arrival rates and charging durations. Additionally, the projected charging demand follows a similar trend to EV arrival rates, peaking during daytime hours. Overall, the study provides insights into managing EV charging infrastructure amidst fluctuating demand and limited capacity.

ABSTRACT. Road infrastructure networks play a vital role in enabling smart mobility, evident in the emergence of innovative transportation concepts aimed at enhancing efficiency and environmental sustainability. This study conducted a comprehensive review of literature focused on road pavement condition assessment, employing computer vision and photogrammetry tools supported by machine learning algorithms to advance smart transportation initiatives. Utilizing the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) declaration, relevant literature from Scopus, Web of Science, Science Direct, Emerald, and ASCI databases were systematically analyzed. Initially identifying 136 articles, the review meticulously filtered down to 56 articles for detailed examination. The review encompasses various aspects, including the utilization of computer vision and photogrammetry tools, model functionalities, infrastructure types, data acquisition methods, and software tools. It also delves into the challenges associated with these techniques, identifies research gaps, and evaluates their potential impact on the integration of autonomous vehicles and smart transportation systems. Key findings emphasize the necessity for standardizing evaluation parameters, ensuring real-time applicability, and considering smart transportation requirements in existing literature. This review underscores the significance of further research in these domains to propel and enhance smart transportation solutions.

ABSTRACT. Velocity measurement based on visual data received from a camera has garnered tremendous attention in the fields of computer vision and image processing. The camera resolution, view angle, and depth stability affect the performance of velocity measurement frameworks. Various descriptors have been developed by researchers; however, they are severely prone to photometric and geometric variations within the image. In this research, we propose an enhanced SIFT local descriptor to extract features from the local structures within the image. Based on feature matching between consecutive frames of the video, we estimated the velocity of a UAV flying at a uniform height above the ground. The proposed descriptor resulted in a more accurate velocity estimate than many state-of-the-art descriptors available in the literature.

ABSTRACT. Smart mobility and logistics (SML) is an emerging paradigm that aims to interconnect transportation networks, industries, and commercial services to provide better services to end-users. SML solutions are rapidly growing in popularity not only in industrial sectors but also at individual levels by means of ride-sharing solutions, micro-mobility solutions, electric vehicles, and intelligent autonomous systems like self-driving cars, trucks, drones, and legged robots, among others. Usually, these solutions utilize communications technologies such as 5G internet services, cloud services, the internet of things, intelligent solutions, sophisticated routing and scheduling algorithms, and security features. The SML sector is still in its early stages, and solutions need to be designed for challenges that are unique due to the diverse needs. Also, many places in the world are facing challenges such as disruption in the global and local supply chains, traffic congestion, the death toll due to poor traffic networks, low-performance routes, and degraded quality of service. This paper aims to help researchers address these challenges and find innovative solutions that can help achieve SML goals.

ABSTRACT. The rising adoption of electric vehicles (EVs) demands the development of efficient and sustainable charging infrastructure. This paper explores designing and optimizing renewable energy-based microgrids tailored for EV charging. It focuses on a case study in Saharanpur City, Uttar Pradesh, India, which benefits from favorable solar irradiance. Through a techno-economic analysis, this study evaluates EV load patterns and the availability of solar and wind resources, ensuring the economic viability of the proposed grid-tied microgrid. A hybrid optimization model is employed to simulate and assess various configurations, emphasizing cost-effectiveness and sustainability. Analysis of 1164 simulated microgrid solutions highlights the critical importance of precise component selection and system design for economic viability, with the net present cost (NPC) ranging between $70.2M and $71.3M. Furthermore, the net financial gain from energy trade costs ranged from $468,726.70 in June to $1,276,238.72 in March. It is evident that the proposed grid-tied microgrid models achieving over 80% renewable fractions demonstrate reduced greenhouse gas emissions and promote economic growth, thereby supporting sustainable energy infrastructure in EV charging.

ABSTRACT. This study delves into the analysis of the information flow topology (IFT) of vehicle convoy communication, specifically comparing two conventional convoy types with a model predictive control (MPC) approach. The research focuses on developing an MPC-based two-vehicle convoy system to control the follower-vehicle in a convoy scenario, considering constraints related to inter-vehicular spacing. The results indicate that the proposed MPC-based system outperformed the conventional convoy types by achieving variable inter-vehicular spacing of 11.7 m and 7.7 m based on driver prediction, enhancing safety and passenger comfort. The controlled vehicle in the proposed model exhibited significant improvements in maintaining variable spacing of 112.0 m,45.0 m and 28.0 m from the lead-vehicle, predecessor-vehicle, and rear-vehicle respectively. Jerk values were reduced to 0.05 〖ms〗^(-3) and 0.02 〖ms〗^(-3) compared to the two conventional convoy setups. By incorporating behavioral prediction into the control strategy, the proposed model demonstrated a more human-like driving behavior, addressing safety concerns and enhancing passenger comfort within the convoy.

ABSTRACT. PEMFCs are a promising technology for mobility applications due to high efficiency, clean emissions and rapid refueling. Developing real-time systems using PEMFCs requires an accurate model with well-defined parameters for design, simulation, and performance evaluation. The paper proposes a new metaheuristic optimization technique, the Newton-Raphson-based optimizer (NRBO) algorithm, for accurate estimation NedStackPS6 fuel cell model parameters. The proposed NRBO algorithm combines ideas from gradient-based and population-based optimization techniques for accurate and efficient parameter estimation. NRBO algorithm bridges the gap between gradient-based and population-based methods. The NRBO algorithm harnesses the exploration efficiency of population-based methods while incorporating gradient information to guide the search towards promising regions within the feasible space. The proposed algorithmic framework for PEMFC parameter estimation is developed based on the Newton-Raphson method for root finding. For accurate estimation of the optimal model parameters for NedStackPS6 PEMFC, NRBO minimizes the total squared error between the estimated and the measured and fuel cell voltage across various data points to find optimal model parameters. NRBO's performance is evaluated by comparing its results to popular optimization algorithms. The statistical comparisons show that NRBO outperforms existing algorithms in accuracy, search capability, and convergence speed.

ABSTRACT. Hydrogen fuel cell trains have emerged as a promising solution for sustainable transportation, offering the potential to reduce greenhouse gas emissions and dependence on fossil fuels. This study conducts a comprehensive examination of the applicability of hydrogen fuel cell trains within the transportation framework, with a specific focus on Saudi Arabia. Through an extensive literature review and rigorous analysis, this research endeavors to elucidate the current state of hydrogen fuel cell train technology and its suitability for various operational contexts. The findings reveal that while the existing technology may not be optimized for heavy freight operations, hydrogen fuel cell trains demonstrate considerable potential for passenger transportation, particularly on shorter routes. Leveraging insights from previous studies and ongoing advancements in hydrogen technology, this paper anticipates a trajectory towards expanded applications of hydrogen fuel cell trains, encompassing heavier and longer-distance operations. Notably, the imminent testing of the first freight train in 2025 underscores the evolving landscape of hydrogen fuel cell train technology. Beyond the confines of Saudi Arabia, this study posits broader implications for the global transportation sector. The analysis underscores the pivotal role of hydrogen fuel cell trains in contributing to decarbonization efforts and fostering sustainable mobility solutions. Moreover, the research identifies opportunities for further exploration and implementation, both within the kingdom and on an international scale. In essence, this research serves as a scholarly inquiry into the feasibility and potential of hydrogen fuel cell trains, laying the groundwork for future academic inquiry, policy development, and technological innovation in the realm of sustainable transportation.

ABSTRACT. The fast depletion of fossil fuels, oil, and natural gas due to their excessive use is a source of motivation for finding alternative ways to overcome their rapid depletion. Global warming is also an important issue caused by the emission of harmful gases. Therefore, various studies have been carried out on Fuel Cell based hybrid electric vehicles to tackle the issue of natural resources depletion and emission of harmful gases. In this paper, the proposed model consists of four sources namely Fuel Cell, solar panel, battery, and supercapacitor. The fuel cell is the primary source whereas the other three sources act as secondary sources. All the sources are connected via DC converters to a DC bus. A double integral sliding mode non-linear controller has been formulated for the hybrid energy storage system. The system’s global stability has been assured through Lyapunov criteria. The proposed model is simulated in MATLAB/Simulink 2024b to verify controller performance. The simulation results authenticate stability, robustness, and efficient performance under various dynamic conditions.

ABSTRACT. Unmanned Aircraft Vehicles (UAVs) are becoming more important in improving healthcare logistics, in particular due to their cost effectiveness, minimized risk, and versatile operational capabilities. This study explores the deployment of autonomous UAVs to deliver medical supplies to remote areas. Advances in ledger technology, smart contracts, and machine learning have transformed tasks previously managed by human teams or manually controlled UAVs into fully autonomous missions. We present a comprehensive analysis of the challenges and initial solutions vital for the effective use of autonomous UAVs in the delivery of medical supplies. In addition, we propose a machine-learning model to optimize UAV performance for these missions.

ABSTRACT. Rapid advancements are being made in autonomous systems for three-dimensional (3D) object identification, which is essential for sensory components. This review paper analyzes cutting-edge 3D object recognition techniques, specifically investigating the integration of Lidar and camera sensors. It also contrasts these techniques with more affordable alternatives, such as utilizing only a camera or combining a camera with Radar. The text emphasizes the limitations of existing techniques, which include significant expenses and practical difficulties, such as the need for real-time data processing and the integration of multiple sensors. The analysis highlights the necessity of implementing creative approaches to tackle these challenges and suggests areas of research to improve the precision of sensors, optimize the integration of data, and develop cost-effective technologies. With this thorough evaluation, our goal is to provide useful insights into the intricacies of 3D object identification, hence promoting future progress in the autonomy of intelligent systems. This work contributes to the continuing discussion on improving the capabilities of autonomous systems by addressing current constraints and investigating potential future opportunities

ABSTRACT. This study examines solar water pumping systems with programmable logic controllers (PLCs) to fill a gap in the literature. PLC-based automation can improve these systems' reliability and efficiency. However, PLC integration with sensors, actuators, and pumps, power consumption optimization, maintenance, and cost-effectiveness prevent their broad implementation. This work analyses two configurations: a DC-DC boost converter-based MPPT solar water pump control scheme and a thorough PLC-based water pumping component research. A well-designed DC-DC boost converter provides motor voltage. This MATLAB-supported method uses solar panels and pumps to solve water scarcity and conserve energy in developing nations

ABSTRACT. This study aims at developing a comprehensive algorithmic UAV autonomous operational toolbox. The toolbox integrates path planning and mission control and accommodates the environmental constraints and external factors of the path domain (natural and manmade obstacle evasion) in addition to the physical and kinematic restraints of a variety of UAVs within a highly defined 3D GIS environment. The developed toolbox is tested on a real-world case study for emergency application, deploying fleets of MEDIC UAVs and assigning the missions for target SOS locations for medical supplies dispatching in an obstacle-rich-environment. For each routing method, the toolbox successfully generated collision-free, near-optimal routes simultaneously. System output includes the route geometry, thrust for each time-step along with the visit order of waypoints for each path. The results show that the route length metric difference is less than 10% with a solution time of less than five seconds. However, comparison of results demonstrate that disturbed fluid paths generate a further stable path that is geometrically simpler for the UAV motion yet marginally lengthier. Practical applications for the proposed toolbox extend beyond emergency response.

ABSTRACT. Deep learning-based image analysis technologies have significantly enhanced decision support systems across various fields, including Geographic Information Systems (GIS) and remote sensing. These technologies are essential in extracting features from remotely sensed images, which are crucial for applications such as land-use planning, transportation infrastructure development, environmental monitoring, etc. Vector shapefiles of buildings can be foundational for various aspects of urban planning, such as land use planning and zoning decisions, infrastructure development, emergency response and management, environmental impact assessment, and many more. Also, detailed shapefiles with accurate geospatial information enhance the efficiency, precision, and public safety in urban development initiatives. Traditional methods of manually digitizing buildings present in an image are not only time-consuming and costly but also less efficient. Despite showcasing promising outcomes, there are several challenges regarding the postprocessing of the output images of deep learning-based models, such as preserving the spatial information embedded in the input images. To deal with this issue, this study introduces a Graphical User Interface (GUI) called EURG DETECTIFY, based on YOLOv8 deep learning model. This GUI streamlines the process of extracting building rooftops from images that have spatial information. This also automatically georeferences the output image by transferring the spatial information from the input image and, lastly, generates precise vector shapefiles of buildings that can be used for visualization and analysis in any GIS software. In summary, EURG DETECTIFY enables spatial information generation through simple, intuitive button interactions.

ABSTRACT. The shift towards electrical transportation system has the promise to reduce greenhouse gas emissions and support sustainability. With the expected rise in electric vehicle (EV) usage in recent years, it becomes critical to develop a charging station infrastructure to ensure the drivers comfortability and optimal performance for EVs. This research recommends that EV drivers need to consider using the combination of the Method (BWM) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to select the most appropriate nearby charging stations based on their personal preferences and specific situations. The primary goal of this study is to improve the process of choosing charging stations based on the drivers’ preferences. By comparing criteria weights and evaluating scores for all nearby charging stations to the drivers. This method seeks to offer a tailored selection process using both available data and user input preferences.

ABSTRACT. In this paper, we propose a crowd shipping application that focuses on using commercial airlines as a mean of transportation to reduce worldwide air-fuel consumption by using the available resources. We achieve this by creating a generic crowd shipping business model that reflects all the domains needed for a successful application. A process flow and two mathematical models were developed: Integer Programming (IP) to ensure the customers-travelers matching and Goal Programming (GP) to investigate the effect of relaxing some the constraints. The work was tested on the North African countries region which are characterized by a variety in travelers, airports, and airport distributions. It was shown that the greater the ratio of travelers to the customers, the fewer deviations were needed and the higher the packages distribution among the travelers is achieved. This study demonstrates that as the network becomes broader with more parties, big data analytics and faster matches are required to secure the scalability of the application.

ABSTRACT. Given the detrimental effects that climate change is having on the ecosystems and general health of our world, we must all work to mitigate or prevent it. Cutting back on carbon emissions is one method we can use to mitigate the effects of climate change. At levels that nature is unable to balance on its own, manufacturing has been a major contributor to the dramatic rise in carbon emissions. In this context, we consider a green capacitated lot-sizing problem in the structure of an industrial symbiosis (GCLS-IS), where two production units (PU1 and PU2) must plan their production over a planning horizon of T periods. PU1 must limit its emissions to a certain government approved CO2 quota in a production period t to satisfy a deterministic time-dependent demand. We assumed that this carbon emission is proportional to the number of units produced per period and can be expressed as a unitary environmental impact. Two types of carbon emission constraints were studied: periodic carbon emission constraints and cumulative carbon emission constraints. Valuable insights derived from this analysis are subsequently used to discuss the managerial implications of setting up the considered industrial symbiosis in terms of model complexity. The periodic and cumulative cases are proven to be NP-hard. Three models were developed; (i) without carbon emissions (ii) periodic carbon emissions (iii) cumulative carbon emission. Upon solving the models with pyomo, model 3 was found to be better than model 2 since it had a lower cost, lower inventory, lower disposal quantity, and did not exceed the approved carbon emissions. Additionally, perspectives to extend this work are discussed.

ABSTRACT. Establish thorough operator training programs before incorporating robotic manipulators into industrial environments. This analysis shows the limitations of standard operator training methods and the need for new ones. Real-life examples show how poor training affects production, safety, and equipment integrity. VR and AR technology have ushered in an interactive and immersive era of operator training. This study examines academic literature and practical implementations to determine industrial VR and AR acceptance and efficacy. This study acknowledges AR-based teaching's budgetary and technological limitations, as well as its potential benefits. Active user interaction, adaptive learning approaches, and authentic simulations are key to successful operator training programs. Additionally, the report discusses recent robotic manipulator research and advances, which can inform future developments and operator training approaches. The study compares VR and AR-based teaching approaches to show their pros and cons. This evaluation covers industrial station robotic manipulator operator training in detail. It underlines the necessity of well-designed training programs in maximizing robotic integration's benefits and establishing the groundwork for future advances in this constantly growing subject

ABSTRACT. This paper provides an in-depth exploration of perception, path planning, and control strategies for autonomous vehicles within the CARLA simulation environment. CARLA, known for its accessibility and open-source nature, serves as the backdrop for this comprehensive study. The research investigates LOAM and Lego-LOAM algorithms using Lidar technology for Simultaneous Localization and Mapping (SLAM) with the goal of improving relocalization efficiency in pre-mapped environments. Moreover, the study integrates global and local path planning techniques to navigate obstacles and select optimal routes. The conventional PID control and the Stanley Method are scrutinized for managing lateral and longitudinal vehicle control,emphasizing their efficacy. These methodologies are seamlessly integrated and rigorously tested within the simulation environment using the Robot Operating System (ROS). The study concludes by highlighting the ease of transitioning proposed perception and control strategies to real-world implementation through ROS. Ultimately, CARLA emerges as a reliable platform for evaluating innovative solutions that drive advancements in the autonomous vehicle industry.

ABSTRACT. In search and rescue operations, time plays a critical role in saving lives. To address this challenge, a multi-drone surveillance system has emerged as a valuable tool for first responders, enabling them to cover large areas efficiently. However, for optimal effectiveness, such a system needs to be collaborative and autonomous, allowing first responders and operational rescue teams to focus on crucial tasks. This paper presents a simulation framework designed to assist in the selection of optimal design characteristics for a multi-drone collaboration system in a specific search and rescue environment, in order to optimize coordination and continuity in large scale missions. The framework takes into account various factors and considerations to guide the decision-making process. By leveraging this simulation framework, stakeholders can evaluate and choose design features that enhance situational insight, optimize resource allocation, and streamline rescue operations in their unique context.

ABSTRACT. Heterogeneous systems integrating ground-mobile vehicle robots and drone UAVs to perform indoor mapping and explore complex operational environments such as industrial areas where the obstacles take places. Despite their potential to enable advanced autonomous exploration with self-reinforcement learning and navigation capabilities, these systems face multiple challenges related to communication, coordination, security, and landscape mapping. This paper discusses the challenges associated with implementing heterogeneous robot systems and examines relevant research articles that contribute to addressing them. Firstly, determining the most effective localization method in unstructured environments, where traditional navigation aids might be limited, poses a significant hurdle. Vision-based approaches for landing the drones on a mobile robot introduce complexities that require innovative solutions. We also need to address the communication challenges that demand real-time and secure data exchanges between vehicular and drone robot systems. Moreover, the limitations of GPS in indoor environments necessitate alternative positioning solutions. Additionally, coordinating leader-follower dynamics between drones and mobile robots requires sophisticated strategies to ensure smooth collaboration and effective mapping. This paper comprehensively examines these challenges and explores relevant research articles that contribute to addressing them, shedding light on potential solutions and avenues for future research.

ABSTRACT. Studying the attenuation and distortion of ultra wideband signals in indoor environment is important for designing wireless com- munication systems. An experimental work was conducted in this paper to characterize the effect of some building materials on ultra wideband propagation. The measurements were taken for the range from 750 MHz to 20 GHz, where this band is used for 6G applications. Pre-processing techniques were applied to remove the effect of noisy bands and multipath effect. The dielectric constant and loss tangent are calculated as a function of frequency for the considered materials.

ABSTRACT. The proliferation of the Internet of Things (IoT) has turned smart homes into a convenience and cyber vulnerability center. We herein present an advanced machine learning framework for the task of intrusion detection in a smart environment. Another unique resource for ML-based cybersecurity applications is provided by the compilation of a large dataset, “Smart Home Intrusion Detection Dataset” that encapsulates diverse attributes of network traffic, specific to smart home IoT devices. This dataset contains all sorts of features, from connection durations to traffic patterns at the granularity of services, which are of utmost importance in distinguishing normal operation from possible security breaches.

Using this data, we formulate a predictive model using various machine learning algorithms and put our efforts into ensemble methods to improve the accuracy and robustness of the model. We rigorously test it on different training and validation sets, leading to the best performance, in turn making it score high for accuracy, precision, recall, and F1-score. We look into model behavior in more detail with the use of a confusion matrix, which refines further the predictive capability of the system.

Our approach is set in the context of strict feature selection, which serves the dual role of giving the best efficiency and making the model more scalable. The results demonstrate the fact that our ML model performs not only better in identifying intrusion attempts but also with fewer false positives, a consideration for practical deployment. This also enhances the strategic defense of smart home ecosystems against an oncoming wave of cyber threats. This further endorses the viability of ML for developing adaptive, responsive intrusion detection systems—a step ahead in security measures for the IoT.

ABSTRACT. Climate change is a well-documented and known phenomenon which is producing multidimensional repercussions around the world. One of them is the increase in frequency of weather events which can have serious impacts on routine activities endangering the sustainability and safety of society. This study aims to investigate one of the impacts of these events which is related to travel behavior. To achieve this goal, the output of an activity-based microsimulation model has been utilized for the Flanders region in Belgium. The information extracted from the model was combined with findings obtained from weather-related travel behaviour studies. Six extreme weather scenarios are defined along with their effects on travel behaviour (especially mode choice). Based on that, the activity-travel schedules of the individuals are revised, readjusted in relation to time and trip distances and aggregate statistics are calculated for comparison with the base case. The analysis also included the determination of CO2 emissions in different scenarios. Results indicate that extreme weather episodes (analyzed in this study) can cause an increase in CO2 emissions by 2 to 6% and therefore, create a vicious cycle. This study also listed some enhancements in traditional bicycle infrastructure improvement strategies to make them more climate-resilient.

ABSTRACT. Control technology and precision navigation advancements have renewed interest in formation flying for commercial and unmanned aerial vehicles (UAVs). Most of the research in this area has focussed on extended formation flying, which is unsuitable and less effective for UAVs. Various bioinspired formation patterns, including the echelon, normal V, and inverted V patterns, have been proposed for research and use in the aviation industry. The trailing UAV should be maintained at an optimal separation distance relative to the lead UAV to maximize the aerodynamic benefits of formation flying. Formation controllers, in conjunction with wake and vortex tracking sensors, can be used to achieve this and maintain the preferred formation pattern and structure. This research examined the effects of close formation flying on the lift, induced drag, and, consequently, the overall range of UAVs. The aerodynamic benefits of flying UAVs in coordinated formation patterns included reduced drag and increased lift by positioning the trailing UAVs at optimal longitudinal, lateral, and vertical separation distances behind the leading UAVs. Through the two examined patterns in this research, the typical V-formation pattern proved better than the inverted V pattern due to the reduced induced adverse rolling effect on the trailing UAV on the former formation pattern.

ABSTRACT. In today's rapidly evolving digital marketplace, the ability to understand and predict market trends and consumer demands using social media analytics is essential. Our study introduces an innovative methodology utilizing the Reformer (Reversible Transformer), an advanced machine learning model that efficiently processes large-scale social media datasets. This model capitalizes on its unique ability to interpret complex data, offering a new perspective on market dynamics as reflected through real-time public sentiment. Our research demonstrates that the Reformer outperforms other models in terms of accuracy, precision, recall, and F1 scores, establishing it as a powerful tool for businesses. By leveraging social media data, companies can obtain crucial market insights and improve strategic decision-making. This study not only validates the Reformer's effectiveness in predictive analytics but also highlights its practical applications in analyzing market trends and forecasting demand. The successful deployment of this methodology marks a significant advancement in the field, empowering businesses to better utilize the wealth of information available on social media platforms to make well-informed decisions. Our approach equips businesses with actionable insights, positioning them to stay competitive in a challenging market environment. The superior performance of the Reformer model underscores its potential as a robust tool for predictive analytics across various real-world applications, making it an invaluable resource for businesses seeking to capitalize on the dynamic nature of digital marketplaces.

ABSTRACT. In the fast-paced e-commerce industry, precise customer demand prediction is essential for sustaining a competitive edge. This study introduces a deep learning-based strategy using the Conditional Transformer Language Model (CTRL) to improve demand forecasting. By leveraging the Amazon Reviews 2018 dataset, the research involves thorough data preprocessing, tokenization, and corpus construction to train sophisticated models. The CTRL model, celebrated for its superior contextual comprehension, undergoes extensive evaluation against conventional machine learning algorithms, with metrics such as accuracy, precision, recall, and the F1 score guiding the identification of the most effective model. The outcomes reveal that CTRL’s nuanced understanding of consumer behavior greatly enhances predictive accuracy. This paper highlights the transformative impact of deep learning in ecommerce, offering businesses crucial insights for proactive market adjustment and strategic inventory management. It underscores the critical role of advanced demand forecasting in securing a market advantage and facilitating informed strategic decisions. By integrating cutting-edge analytical techniques, this research not only boosts predictive performance but also provides a robust framework for e-commerce businesses aiming for sustainable growth and improved operational efficiency. This study exemplifies how deep learning can revolutionize market adaptability and business strategy in the dynamic world of e-commerce.

ABSTRACT. Road networks are significant and instrumental assets for efficiently and reliably connecting last-mile distribution centers to end consumers. They play a crucial role in facilitating the final leg of the supply chain, ensuring timely deliveries and customer satisfaction. Road cracks and potholes on road networks pose significant challenges because they impede the smooth flow of goods and vehicles. These infrastructure issues can lead to delays, vehicle damage, and safety concerns, thereby increasing transportation and human costs. Delivery services between distribution centers and end users have recently utilized autonomous trucks and other delivery vehicles. Autonomous vehicle navigation systems need to be equipped with artificial intelligence models to timely detect and maneuver to avoid potential damage to the vehicle as well as the goods they are transporting. The deployment of autonomous delivery vehicles in metropolitan cities characterized by damaged roads poses a unique challenge. To ensure the safe and reliable provision of logistics services, autonomous vehicles must detect and avoid potholes and road cracks. In this paper, our research focuses on the critical task of object detection, with a specific emphasis on identifying potholes and road cracks, so that companies providing Mobility as a Service (MaaS) or autonomous vehicle companies may integrate object detection models into their systems. Leveraging advanced deep learning techniques, we propose a novel approach that utilizes an optimized object detection algorithm with spatial analysis to accurately detect and classify potholes and road cracks in real-time. In this study, we utilized the latest deep learning You Look Only Once (YOLO) models and optimized them for better results using hyperparameter techniques. To compare the performance of these optimized models, the mean average precision (mAP) at 50% intersection over union (mAP50) metric was selected. The results indicated the efficacy of the trained models in accurately identifying road irregularities amidst the complex road surface structure. Our findings have significant implications for the logistics industry, allowing carriers to employ autonomous vehicles to optimize delivery operations by detecting road irregularities and taking appropriate decisions to avoid them.

ABSTRACT. This paper investigates the dynamics of a hybrid power system (HPS) and powertrain configurations in a fuel cell electric vehicle (FCEV), emphasizing its nonlinear behavior. The study focuses on a hybrid powertrain setup comprising a fuel cell (FC) as the primary source, complemented by an ultracapacitor (UC) and a battery as auxiliary sources, each subject to uncertainties. The system incorporates DC/DC boost converters linking the FC to the DC-Link, and bi-directional converters connecting the UC and battery to the DC-Link. Dynamic models are developed to capture the nonlinear characteristics of the subsystems. The primary goal is to develop a controller capable of simultaneously managing the DC/DC converters amidst uncertainties. A nonlinear control strategy based on Lyapunov theory is proposed to regulate the DC-link voltage and accurately track currents from the sources. The study ensures closed-loop stability under different operating conditions and uncertainties. Simulation results in MATLAB validate the efficacy of the proposed controllers, justifying the UC and battery's ability to meet load demands and enhance fuel cell performance.