Sensoriamento urbano e cidades inteligentes têm sido tópicos derivados da computação ubíqua em alta nos últimos anos, tanto para a academia como para a indústria, devido ao contínuo avanço tecnológico aliado à maior facilidade de acesso e aceitação pelos usuários. Na literatura pesquisada sobre plataformas que englobam tais tópicos foi constatado que diversas delas possuem algum processo autonômico utilizado para atender alguma necessidade de autoadaptação em tempo de execução. Apesar disso, nenhuma das plataformas pesquisadas focou especificamente em encontrar e propor uma solução para tratar exclusivamente a autoadaptação. Nesse contexto, esta dissertação tem por objetivo propor um mecanismo de autoadaptação para sistemas de sensoriamento urbano, além de avaliar seu comportamento. Como primeiro passo para realizar tal objetivo, foi conduzida uma pesquisa literária tendo em vistas identificar os principais casos de adaptação em sistemas de sensoriamento urbano, além de requisitos específicos da arquitetura de sensoriamento urbano UrboSenti, utilizada para implementação. Como segundo passo, a partir dos requisitos identificados, o modelo MAPE-K da computação autonômica foi escolhido como a base da construção do mecanismo de autoadaptação. A implementação deste modelo utilizou as técnicas de eventos passivos para monitoramento do ambiente, regras Evento-Condição-Ação, para tomada de decisão, planos estáticos para planejamento e adaptações por parâmetros e componentes para execução. Tanto o modelo como as técnicas escolhidas foram implementadas devido atenderem as necessidades dos cenários avaliados. Por fim, as avaliações aplicadas apontam resultados preliminares satisfatórios, dados os casos avaliados e os experimentos de tempo de resposta a eventos internos e interações; no entanto, tais avaliações revelarem diversos pontos que devem ser explorados em trabalhos futuros. / In the last years, urban sensing and smart cities have been popular topics derived from the ubiquitous computing, for both the academia and the industry, due to its continuous technological development combined with greater facilities of access and acceptance by the users. The reviewed literature about platforms that encompass such topics showed that many of them have some kind of autonomic process used to meet any need for self-adaptation at runtime. Despite this, none of the researched platforms focused in proposing a solution to exclusively meet the self-adaptation properties. In this way, this dissertation aims to propose a self-adaptive mechanism to urban sensing systems, as well as evaluating its behavior. As the first step to achieving such goal, a literature review was performed aiming to identify the main adaptation cases in urban sensing systems, as well to identify the specific requirements of the UrboSenti architecture for urban sensing. As the second step, the autonomic computing MAPE-K model was chosen to compose the foundation of the self-adaptive mechanism based on the identified requirements. The implementation of this model used the techniques of passive events for monitoring, rules Event-Condition-Action for decision making, static plans for planning and parameter and component adaptations for execution were used in the proposed implementation to meet the evaluated scenario needs. Lastly, the applied evaluations indicate satisfactory results, given the assessed cases and the experiments of scalability at the response of internal events and interactions. However, they have left many open points that should be explored in future works.

A urbanização acentuada, fenômeno crescente das últimas décadas, tem atingido níveis elevadíssimos e criado enormes desafios para a gestão das cidades, além de trazer uma vasta gama de efeitos nefastos para a qualidade de vida de seus cidadãos. Dados da ONU de 2016, indicam tratar-se de um caminho sem volta, com uma tendência de agravamento nos próximos anos. Para tentar mitigar esta situação, muito se tem discutido em como aumentar o nível de inteligência das cidades e o interesse pelo tema Cidades Inteligentes tem crescido. Apesar disso, ainda não existe consenso sobre um conceito de cidade inteligente. Se há tempos este conceito se baseava exclusivamente no pilar da tecnologia, hoje uma visão mais evoluída e holística incorpora várias outras dimensões. Já a maioria dos modelos de classificação existentes são estrangeiros e não são aderentes à realidade de um país tão diverso e tão desigual quanto o Brasil. No âmbito nacional, existem algumas iniciativas de criação de conceitos e de modelos de classificação, mas, ou são baseados unicamente no componente tecnológico, ou buscam apenas a criação de rankings tradicionais baseados em ponderações arbitrárias de seus formuladores. A dificuldade de adequação do conceito e dos modelos de classificação à realidade brasileira foram os dois motores principais desta tese. O primeiro objetivo foi desenvolver um conceito de cidade inteligente para o contexto brasileiro. O segundo foi propor um modelo multidimensional de classificação, que fugisse aos padrões tradicionais de um ranking e fosse um instrumento efetivo de aprendizagem, benchmarking e de apoio ao planejamento de políticas públicas das cidades. Por meio de um estudo exploratório e descritivo, junto a 3 cidades de portes diferentes do estado de São Paulo, foram desenvolvidas pesquisas de abordagem quantitativa e qualitativa. A primeira utilizou questionários fechados e levantamento de dados em bases de indicadores específicos para o cálculo dos componentes do Índice Brasileiro Multidimensional de Classificação de Cidades Inteligentes - IBMCCI. Este produto final da tese emprega a mesma abordagem orientada ao usuário do U-MULTIRANK, o ranking multidimensional global de universidades da Comunidade Europeia. A versão final do modelo proposto foi disponibilizada para uso dos gestores municipais com a liberdade de seleção das dimensões, dos indicadores e dos municípios equivalentes para análise. A abordagem qualitativa da pesquisa foi conduzida por meio de entrevistas semiestruturadas, junto a 2 especialistas em gestão municipal. Para validar a ferramenta construída utilizou-se a técnica de validação de conteúdo. Chegou-se à conclusão que é necessário ajustar alguns fatores do modelo e adequar a periodicidade de edição do índice para coincidir com o calendário das eleições municipais. Com estas adequações, o IBMCCI demonstrou grande potencial de se tornar uma ferramenta efetiva de apoio para os formuladores de políticas públicas municipais. / The strong urbanization, a phenomenon that has increased in recent decades, has reached a very high level and has created enormous challenges for the management of cities, as well as bringing a wide range of harmful effects to the quality of life of its citizens. UN data of 2016 indicates that this is one-way road, and it\'s forecast to be worse in the coming years. Trying to mitigate this situation, much has been discussed about how to increase the smartness level of the cities and the interest in subject of Smart Cities has grown. Despite this, there is still no consensus on a smart city concept. If for some time this concept was based exclusively on the pillar of technology, today a more evolved and holistic vision would incorporate several other dimensions. Most of the existing classification models are foreign and are not adherent to the reality of a country as diverse and as unequal as Brazil. At the national level, there are some initiatives to create concepts and classification models, but they are based solely on the technological component or they only seek to create traditional rankings based on arbitrary weights of their formulators. The difficulty of adapting the concepts and the classification models to the Brazilian reality were the two main engines of this thesis. The first aim was to develop a concept of Smart cities for the Brazilian context. The second one was to propose a multidimensional classification model, which would escape the traditional standards of a ranking and be an effective tool for learning, benchmarking and supporting the planning of public policies in cities. By means of an exploratory and descriptive study, along with 3 cities of different sizes of the state of São Paulo, research into quantitative and qualitative approaches were developed. The first one used close ended questionnaires and data collection based on specific indicators for the calculation of the components of the Brazilian Multidimensional Smart Cities Classification Index (IBMCCI). This final thesis product employs the same user-oriented approach as U-MULTIRANK, the multidimensional global ranking of European Community universities. The final version of the proposed model was made available for use by municipal managers with the freedom to select dimensions, indicators and equivalent municipalities for analysis. The qualitative approach of the research was conducted through semi-structured interviews, with 2 specialists in municipal management. In order to validate the built tool, the content validity technique was used. It has come to the conclusion that it is necessary to adjust some factors of the model and to adapt the periodicity of edition of the index to coincide with the calendar of the municipal elections. With these adaptations, the IBMCCI has demonstrated great potential to become an effective support tool for municipal public policy makers.

Smart City technologies emerge as a potential solution to tackle common problems in large urban centers by using city resources efficiently and providing quality services for citizens. Despite the various advances in middleware technologies to support future smart cities, there are yet no widely accepted platforms. Most of the existing solutions do not provide the required flexibility to be shared across cities. Moreover, the extensive use and development of non-open-source software leads to interoperability issues and limits the collaboration among R&D groups. Our research explores the use of a microservices architecture to address key practical challenges in smart city platforms. More specifically, we are concerned with the impact of microservices on addressing the key non-functional requirements to enable the development of smart cities such as supporting different scalability demands and providing a flexible architecture which can easily evolve over time. To this end, we are developing InterSCity, a microservice-based open source smart city platform that aims at supporting the development of sophisticated, cross- domain applications and services. Our early experience shows that microservices can be properly used as building blocks to achieve a loosely coupled, flexible architecture. Experimental results point towards the applicability of our approach in the context of smart cities since the platform can support multiple scalability demands. We expect to enable collaborative, novel smart city research, development, and deployment initiatives through the InterSCity platform. The full validation of the platform will be conducted using different smart city scenarios and workloads. Future work comprises the ongoing design and development effort on data processing services as well as more comprehensive evaluation of the proposed platform through scalability experiments. / As tecnologias de Cidades Inteligentes surgem como uma potencial solução para lidar com problemas comuns em grandes centros urbanos, utilizando os recursos da cidade de maneira eficiente e fornecendo serviços de qualidade para os cidadãos. Apesar dos vários avanços nas tecnologias de middleware para suporte às cidades inteligentes do futuro, ainda não existem plataformas amplamente aceitas. A maioria das soluções existentes não oferece a flexibilidade necessária para ser compartilhada entre as cidades. Além disso, o vasto uso e desenvolvimento de software proprietário levam a problemas de interoperabilidade e limitam a colaboração entre grupos de P&D. Nesta dissertação, exploramos uso de uma arquitetura de microsserviços para abordar os principais desafios práticos em plataformas de cidades inteligentes. Mais especificamente, estamos preocupados com o impacto dos microsserviços sobre requisitos não-funcionais para permitir o desenvolvimento de cidades inteligentes, tais como o suporte a diferentes demandas de escalabilidade e o fornecimento de uma arquitetura flexível que pode evoluir facilmente. Para esse fim, criamos a InterSCity, uma plataforma para cidades inteligentes de código aberto baseada em microsserviços que visa apoiar o desenvolvimento de aplicativos e serviços sofisticados em múltiplos domínios. Nossa experiência inicial mostra que os microsserviços podem ser usados adequadamente como blocos de construção para obter uma arquitetura flexível e fracamente acoplada. Resultados experimentais apontam para a aplicabilidade de nossa abordagem no contexto de cidades inteligentes, já que a plataforma pode suportar diferentes demandas de escalabilidade. Esperamos permitir pesquisas colaborativas e inovadoras em cidades inteligentes, assim como o desenvolvimento e iniciativas de implantações reais através da plataforma InterSCity. A validação completa da plataforma será realizada usando diferentes cenários de cidades inteligentes e cargas de trabalho. Os trabalhos futuros compreendem o esforço contínuo de projetar e desenvolver novos serviços de processamento de dados, bem como a realização de avaliações mais abrangentes da plataforma proposta por meio de experimentos de escalabilidade.

Memoria para optar al título de Ingeniero Civil Eléctrico / Una Smart City es un concepto asociado a un tipo de desarrollo urbano basado en la sostenibilidad, desarrollo que es capaz de responder a las necesidades básicas de instituciones, empresas y de los propios habitantes, tanto en el plano económico como en los aspectos operativos, sociales y ambientales. El rol de las TIC es pasivo, remitiéndose a la recopilación y análisis de datos, y la optimización en la utilización de infraestructuras, así como facilitar la comunicación entre los diferentes servicios de la ciudad. Asimismo, la introducción de nuevas tecnologías emergentes se convierte en una oportunidad por la cual diversas comunidades políticamente organizadas optan por adoptar, gracias a los beneficios energéticos que estas entregan, en conjunto con la inteligencia que estas proporcionan. De esta manera, múltiples ciudades han decidido integrar infraestructuras TICAR (Tecnología de Información, Telecomunicaciones, Automatización y Robótica) a fin de responder adecuadamente a diversas necesidades. En el caso de Chile, si bien existen diversas iniciativas para apoyar la investigación, desarrollo e integración de nuevas tecnologías, estas no logran ser suficientes para el potencial mismo del país. En el contexto de Smart Cities, el país presenta un gran potencial para integrar tecnologías TICAR, gracias al sólido conocimiento existente de las mismas, la factibilidad económica para su adquisición, y la existencia de múltiples infraestructuras. Referido a este último punto, una infraestructura de gran interés resulta ser el alumbrado público. Este sistema se caracteriza por iluminar zonas urbanas y sectores residenciales, con el objetivo de proporcionar la visibilidad adecuada para el normal desarrollo de actividades. Si existe una infraestructura física, que dispone de recursos energéticos, ¿por qué no diseñar un sistema de luminarias inteligentes, que responda como infraestructura habilitante para Smart Cities, con versatilidad para la integración de diversas tecnologías de comunicaciones e IoT? En este contexto, la realización del presente trabajo de título tiene por objetivo modelar una infraestructura TICAR habilitante para Smart Cities, con foco a las redes de comunicaciones. Como infraestructura habilitante se propone el desarrollo de una red pública de luminarias inteligentes. Para esto, se realiza una revisión profunda de antecedentes que permitan contextualizar la problemática presente, una comparación preliminar entre tecnologías IoT que permita observar el desempeño y comportamiento de las mismas en entornos urbanos, y el diseño de la infraestructura habilitante desde la perspectiva de las tecnologías de la comunicación. A fin de validar esta propuesta, se considera dentro de los parámetros de diseño estandarizaciones y regulaciones de ambientes urbanos y tecnologías emergentes, junto a la utilización de plataformas de simulación que permitan caracterizar de manera robusta cada uno de los escenarios elaborados.

In the domain of social design, this thesis introduces a design project that explores a place-centric view on interaction design to address issues of participation, representation, and place. The project aims to envision possible futures of city-scaled digital twins as places that enable collective engagement within communities and between citizens and cities. It explores this aim with a methodology grounded in co-design- and ethnographic values, focusing on the place of Ringön in Gothenburg. The final design outcome of the study is an augmented reality-based application connected to the city-scaled digital twin of Gothenburg. The application introduces a concept that enables citizens to express location-based social and cultural values and engage in common concerns. It enables the City of Gothenburg to communicate with citizens and develop an enriched understanding of the identity and character of places in Gothenburg. The study hopes to contribute with knowledge to the body of work within social design that explores how sustainable smart cities can develop from a bottom-up perspective, in a more participatory and socially sustainable way.

This thesis summarises the outcomes of a Big Data analysis, performed on a set of hourly district heating energy consumption data from 2012 for nearly 15 000 buildings in the City of Stockholm. The aim of the study was to find patterns and inefficiencies in the consumption data using KNIME, a big data analysis tool, and to initiate a retrofitting plan for the city to counteract these inefficiencies. By defining a number of energy saving scenarios, the potential for increased efficiency is estimated and the resulting methodology can be used by other (smart) cities and policy makers to estimate savings potential elsewhere. In addition, the influence of weather circumstances, building location and building types is studied. In the introduction, a concise overview of the concepts Smart City and Big Data is given, together with their relevance for the energy challenges of the 21st century. Thereafter, a summary of the previous studies at the foundation of this research and a brief theory review of less common methods used in this thesis are presented. The method of this thesis consisted of first understanding and describing the dataset using descriptive statistics, studying the annual fluctuations in energy consumption and clustering all consumer groups per building class according to total consumption, consumption intensity and time of consumption. After these descriptive steps, a more analytical part starts with the definition of a number of energy saving scenarios. They are used to estimate the maximal potential for energy savings, regardless of actual measures, financial or temporal aspects. This hypothetical simulation is supplemented with a more realistic retrofitting plan that explores the feasibility of Stockholm’s Climate Action Plan for 2012-2015, using a limited set of energy efficiency measures and a fixed investment horizon. The analytical part is concluded with a spatial regression that sets out to determine the influence of wind velocity and temperature in different parts of Stockholm. The conclusions of this thesis are that the potential for energy savings in the studied data set can go up to 59% or 4.6 TWh. The financially justified savings are estimated at ca. 6% using favourable investment parameters. However, these savings quickly diminish because of a high sensitivity on the input parameters. The clustering analysis has not yielded the anticipated results, but they can be used as a tool to target investments towards groups of buildings that have a high return on investment.

The application of Internet of Things (IoT) where sensors and actuators embedded in physical objects are linked through wired and wireless networks has shown a rapid growth over the past years in various domains with the benefits of improving efficiency and productivity, reducing cost, providing mobility and agility, etc. This dissertation focuses on developing signal processing and machine learning based techniques in IoT with applications to 1) smart energy generation and 2) robust indoor localization in smart city. Smart grids, in contrast to legacy grids, facilitate more efficient electricity generation and consumption by allowing two-way information exchange among various components in the grid and the users based on the measurements from numerous sensors located at different places. Due to the introduction of information communications, a smart grid is faced with the risk of external attacks which is aimed to take control of the grid. In particular, electricity generation from photovoltaic (PV) systems is a mature power generation technology utilizing renewable resources, owning to its advantages in clean production, reduced cost and high flexibility. However, the performance of a PV system can be susceptible and unstable due to various physical failures and dynamic environments (internal circuit faults, partial shading, etc.). To safeguard the system security, fault or attack detection technologies are of great importance for PV systems and smart grids. Existing approaches on fault or attack detection either rely on the prediction by a predetermined system model which acts as reference data for comparison or can be applied only within a certain set of component (e.g., several PV strings) based on local statistical properties without the capability of generalization. Furthermore, the output performance of a PV system is dynamic under different environmental conditions (irradiance level, temperature, etc.), which can be optimized by the technique of maximum power point tracking (MPPT). However, previous studies on MPPT usually require prior knowledge of the system model or high computational complexity for iterative optimization. Smart city, as another important application of IoT, relies on analysis of the measurement data from sensors located at users and environments to provider intelligent solutions in our daily life. One of the fundamental tasks for advanced location-based services is to accurately localize the user in a certain environment, e.g., on a certain floor inside a building. Indoor localization is faced with challenges of moving users, limited availability of sensors and noisy measurements due to hardware constraints and external interferences. This dissertation first describes our advanced fault/attack detection and localization methods for PV systems and smart grids, then develops our enhanced MPPT techniques for PV systems, and finally presents our robust indoor localization methods for smartphone users, based on statistical signal processing and machine learning approaches. In Chapter 2 and Chapter 3, we proposes fault/attack detection method in PV systems and smart grids respectively in the framework of abrupt change detection utilizing sequential output measurements without assuming any prior knowledge of the system characteristics or particular faulty/attack patterns, such that an alarm will triggered regardless of the magnitude or the type of faulty/attack signals. Starting from the proposed fault detection method in Chapter 2, we present our fault localization method for PV systems in Chapter 4 where the central controller is able to identify the faulty PV strings without full knowledge of each local measurements. Chapter 5 studies the MPPT method under dynamic shading conditions where we adopt neural networks to assist the identification of the global maximum power point by depicting the relationship between the system output power and the operating voltage. In Chapter 6, to tackle the challenge of accurate and robust indoor localization for smart city when sensors provides noisy measurement data, we propose a cooperative localization method which exploits the readings of the received strengths of Wi-Fi signals at the smartphone users and the relative distances among neighboring users to combat the deterioration due to aggregated measurement errors. Throughout the dissertation, our proposed methods are followed by simulations (of a PV system or a grid under various operating conditions) or experiments (of localizing moving users with smartphones to record sensors' measurements). The results demonstrate that our proposed fault/attack detection and localization methods and MPPT schemes can achieve higher adaptivity and efficiency with robustness against various external conditions an lower computational complexity, and our cooperative localization methods have high localization accuracy even given large measurement errors and limited measurement data.

Magister Commercii (Information Management) - MCom(IM) / The advent of the 4th Industrial Revolution brought on an onslaught of technology rippling through a multitude of industries. Smart Cities, Smart Communities, Artificial Intelligence and Cloud Computing are but a few buzzwords of this digital age. It is argued in Information Systems that many of the challenges faced by communities can be addressed in part through the innovative use of technology. As Higher Education (HE) communities move from traditional campus communities to smart campus communities, the application and implementation of technological advancements and digital skills are needed to facilitate the transition. The disruption caused by COVID-19 virus has had a significant effect on the tertiary educational sector. This research is particularly important and relevant in a post-pandemic phase in which HE finds itself. Especially, establishing a technological and digitally equipped HE community to safeguard itself from possible future threats that impede daily operations within HE campus communities.

Autonomously monitoring and analyzing the behavior of the crowd is an open research topic in the transportation field because of its criticality to the safety of people. Real-time identification, tracking, and prediction of crowd behavior are primordial to ensure smooth crowd management operations and the welfare of the public in many public areas, such as public transport stations and streets. This being said, enabling such systems is not a straightforward procedure. First, the complexity brought by the interaction and fusion from individual to group needs to be assessed and analyzed. Second, the classification of these actions might be useful in identifying danger and avoiding any undesirable consequences. The adoption of the Internet-of-things (IoT) in such systems has made it possible to gather a large amount of data. However, it raises diverse compatibility and trustworthiness challenges, among others, hindering the use of conventional service discovery and network navigability processes for enabling crowd management systems. In fact, as the IoT network is known for its highly dynamic topology and frequently changing characteristics (e.g., the devices' status, such as availability, battery capacity, and memory usage), traditional methods fail to learn and understand the evolving behavior of the network so as to enable real-time and context-aware service discovery to assign and select relevant IoT devices for monitoring and managing the crowd. In large-scale IoT networks, crowd management systems usually collect large data streams of images from different heterogeneous sources (e.g., CCTVs, IoT devices, or people with their smartphones) in an inadvertent way. Due to the limitations and challenges related to communication bandwidth, storage, and processing capabilities, it is unwise to transfer unselectively all the collected images since some of these images either contain duplicate information, are inaccurate, or might be falsely submitted by end-users; hence, a filtering and quality check mechanism must be put in place. As images can only provide limited information about the crowd by capturing only a snapshot of the scene at a specific point in time with limited context, an extension to deal with videos to enable efficient analysis such as crowd tracking and identification is essential for the success of crowd management systems. In this thesis, we propose to design a smart image enhancement and quality control system for resource pooling and allocation in the Internet-of-Things applied to crowd management systems. We first rely on the Social IoT (SIoT) concept, which defines the relationships among the connected objects, to extract accurate information about the network and enable trustworthy and context-aware service exchange and resource allocation. We investigate the service discovery process in SIoT networks and essentially focus on graph-based techniques while overviewing their utilization in SIoT and discussing their advantages. We also propose an alternative to these scalable methods by introducing a low-complexity context-aware Graph Neural Network (GNN) approach to enable rapid and dynamic service discovery in a large-scale heterogeneous IoT network to enable efficient crowd management systems. Secondly, we propose to design a smart image selection procedure using an asymmetric multi-modal neural network autoencoder to select a subset of photos with high utility coverage for multiple incoming streams in the IoT network. The proposed architecture enables the selection of high-context data from an evolving picture stream and ensures relevance while discarding images that are irrelevant or falsely submitted by smartphones, for example. The approach uses the photo's metadata, such as geolocation and timestamps, along with the pictures' semantics to decide which photos can be submitted and which ones must be discarded. To extend our framework beyond just images and deal with real-time videos, we propose a transformer-based crowd management monitoring framework called V3Trans-Crowd that captures information from video data and extracts meaningful output to categorize the crowd's behavior. The proposed 3D Video Transformer is inspired from Video Swin-Transformer/VIVIT and provides an improved hierarchical transformer for multi-modal tasks with spatial and temporal fusion layers. Our simulations show that due to its ability to embed the devices' features and relations, the GNN is capable of providing more concise clusters compared to traditional techniques, allowing for better IoT network learning and understanding. Moreover, we show that the GNN approach speeds up the service lookup search space and outperforms the traditional graph-based techniques to select suitable IoT devices for reporting and monitoring. Simulation results for three different multi-modal autoencoder architectures indicate that a hierarchical asymmetric autoencoder approach can yield better results, outperforming the mixed asymmetric autoencoder and a concatenated input autoencoder, while leveraging user-side rendering to reduce bandwidth consumption and computational overhead. Also, performance evaluation for the proposed V3Trans-Crowd model has shown great results in terms of accuracy for crowd behavior classification compared to state-of-the-art methods such as C3D pre-trained, I3D pre-trained, and ResNet 3D pre-trained on the Crowd-11 and MED datasets.

<p dir="ltr">The roads in the US received a "D" grade on the 2021 report card for America's infrastructure by the American Society of Civil Engineers (ASCE). Poor road conditions generally translate into traffic accidents and vehicle damage, which result in additional expenses for drivers in terms of vehicle repairs and operating costs. To maintain a satisfactory pavement condition over an extended period of time, frequent inspections should be conducted, and any existing and imminent defects should be promptly addressed through corrective and preventive maintenance. However, the current practices are hindered by issues of inspectors' subjectivity, delayed responsiveness, and high costs. This study aims to develop innovative solutions that harness Artificial Intelligence (AI), robotics, and virtual reality (VR) to enhance pavement quality in smart cities.</p><p dir="ltr">The study developed an autonomous system that relies on crowdsourced RGB and depth (RGB-D) data to assess road conditions. A cost-effective data acquisition system that can be mounted on multiple vehicles, was developed. Armed with a substantial dataset of RGB-D pavement surface data, this study explores the effectiveness of various depth-encoding techniques and RGB-D data fusion methods, using pothole detection as a case study. Comprehensive experiments were conducted to evaluate the effectiveness of defect detection using deep convolutional neural networks (DCNN). This study considered all major types of pavement defects in order to comprehensively evaluate pavement conditions. The Pavement Surface Evaluation Rating (PASER) for asphalt pavement is used as a case study. The establishment of an expert system for pavement condition evaluation involves the classification and quantification of pavement data. The system also facilitates the tracking of identified defects and repair work, providing up-to-date information on pavement deterioration and maintenance.</p><p dir="ltr">Another aspect of this study is the improvement of pavement maintenance quality. To enhance the assessment of the effectiveness of pavement maintenance, this study developed immersive VR modules that provide technical staff with a supplementary platform for training. The training materials focus on two common types of pavement maintenance operations: crack sealing and patching. These modules include an interactive decision-making module for evaluating the quality of operations, as well as a hands-on task-performing module for crack sealing machinery preparation and the procedure of full-depth patching. This dissertation has revealed innovative approaches for integrating cutting-edge technologies into the assessment of pavement conditions. The proposed research aims to improve the safety, </p>