Trends in Cyber-Physical Multi-Agent Systems. The PAAMS Collection - 15th International Conference, PAAMS 2017

In this paper a network-oriented computational model is presented for voting intentions over time specifically for the race between Donald Trump and Hillary Clinton in the 2016 US presidential election. The focus was on the role of social and mass communication media and the statements made by Donald Trump or Hillary Clinton during their speeches. The aim was to investigate the influence on the voting intentions and the final voting. Sentiment analysis was performed to check whether the statements were high or low in language intensity. Simulation experiments using parameter tuning were compared to real world data (3 election polls until the 8th of November).

This study describes the use of adaptive temporal-causal networks to model and simulate the development of mutually interacting opinion states and connections between individuals in social networks. The focus is on adaptive networks combining the homophily principle with the more becomes more principle. The model has been used to analyse a data set concerning opinions about the use of alcohol and tobacco, and friendship relations. The achieved results provide insights in the potential of the approach.

In order to increase sales and profits, it is common that e-commerce website owners resort to several marketing and advertising techniques, attempting to influence user actions. Summarizing and analysing user behaviour is a complex task since it is hard to extrapolate patterns that never occurred before and the causality aspects of the system are not usually taken into consideration. There has been studies about characterizing user behaviour and interactions in e-commerce websites that could be used to improve this process. This paper presents an agent-based framework for simulating models of user behaviour created through data mining processes within an e-commerce context. The purpose of framework is to study the reaction of user to stimuli that influence their actions while navigating the website. Furthermore a scalability analysis is performed on a case-study.

The aim of this article is to present home automation system which can be easily integrated into existing buildings without expensive cable installation. A novel solution of a Smart Home control system is proposed and described. The system is based on a number of mutually interconnected nodes, where each node is fully autonomous and can react to user requests directly. Any sensor value or change state of any connected actuator with regards to its position in the system can be measured. Users can interact with the system using web interface situated in one of the nodes. This node then interprets the commands to other devices in the system. The master node owns a list of interconnected devices including their unique identifiers and network addresses. Commands are sent between nodes using IP protocols and using TCP sockets so all responses almost runs in real time.The node’s autonomy is very useful, if there is no Internet connection available. Thanks to their autonomy, nodes can also independently perform specified activities like long term data collection, and react to the current state of monitored environment using internal rules and schedules.

It is well-known that the high penetration of Distributed Energy Resources (DERs) can be troublesome because of their unpredictable behaviors. In this context, renewable energy source (RES) appears as one of the most random component, since, in general, they are weather-dependent. The present paper develops a methodology to evaluate the reliability impacts of RES penetration in microgrid’s distribution system. For this evaluation, was used the Colored Petri Nets (CPN) formalism and event-driven analysis, in order to formulate the stochastic behaviors and simulate the system. To avoid the complexity on modeling a microgrid, the agent approach was used, which permits to manage each component as a unique entity, and assemble the whole system using the communication between the agents. Concerning the validation of the proposed methodology, a comparison between the results of CPN modeling and Monte Carlo simulation is done by means of statistical analysis.

The increase of distributed generation in several countries has led to new legislation that allows the owners to use the energy obtained from them in three possible ways: use the energy to face own consumption such as on-site generation, sell energy to the grid as a producer, or finally, do both according to context of operation. In this way, these technologies can be more easily introduced to the average public if there is a managing application that can represent the interests of its owners and perform the appropriate measures. This paper proposes a methodology for the management of different available technologies owned by a prosumer, analyzing the possible role it can have and what type of scheduling can be made, operating in the third way mentioned before.

Electric power systems have been altered at the operating and planning level in the last years. Evidence of this was the liberalization of the energy market and the implementation of concepts such as smart grids and the active participation of consumers through demand response programs. The main objective of the present paper is to propose an emulation model for asynchronous motors, in the scope of dynamic measurement and control of loads, included in demand response programs. Since this type of electric motor is of upmost importance in the current consumption context, it was necessary to model consumption and evaluate its impact on the power system. The development of this paper aims to evaluate the system’s capabilities using the motor as a load, and a wind emulator. Thus, we can represent various applications (e.g. division of a house) and realize the effect of distributed generation in the operation of the electric power system.

This paper proposes a Predictive Dispatch System (PDS) as part of a Multi-Agent system that models the Smart Home Electricity System (MASHES). The proposed PDS consists of a Decision-Making System (DMS) and a Prediction Engine (PE). The considered Smart Home Electricity System (SHES) consists of different agents, each with different tasks in the system. A Modified Stochastic Predicted Bands (MSPB) interval optimization method is used to model the uncertainty in the Home Energy Management (HEM) problem. Moreover, the proposed method to solve HEM problem is based on the Moving Window Algorithm (MWA). The performance of the proposed Home Energy Management System (HEMS) is evaluated using a JADE implementation of the MASHES.

This paper presents the demonstration of an energy resources management approach using a physical smart city model environment. Several factors from the industry, governments and society are creating the demand for smart cities. In this scope, smart grids focus on the intelligent management of energy resources in a way that the use of energy from renewable sources can be maximized, and that the final consumers can feel the positive effects of less expensive (and pollutant) energy sources, namely in their energy bills. A large amount of work is being developed in the energy resources management domain, but an effective and realistic experimentation are still missing. This work thus presents an innovative means to enable a realistic, physical, experimentation of the impacts of novel energy resource management models, without affecting consumers. This is done by using a physical smart city model, which includes several consumers, generation units, and electric vehicles.

This paper has as main objective the performance evaluation of the Gravitational Search Algorithm for Demand Response programs applied to residential consumers. For this purpose, it was considered a model that describes the consumption and energy price, according to the loads present in a residence. This way, it is intended to minimize the cost of electricity for final consumers based on an optimized planning of their loads at different times. In addition, it will be considered a variable cost of electricity over time (hourly price). In this sense, the cost of electricity will be discretized throughout the day. Finally, the performance of the Gravitational Search Algorithm for the considered model will be evaluated.

Measuring and recording systems for the consumption of electrical energy which are connected to households, are essential in the optimization of energy use. Non-Intrusive Load Monitoring (NILM) is one of the most used techniques in the study of electrical consumption; these systems are based on the analysis of the load curve (the aggregated electrical consumption of the whole household). Thanks to a significant reduction in the price of sensors and sensor systems in recent years, it is possible to individually monitor each one of the devices connected to the grid. In this paper we compare different classifiers in order to find out which is the most appropriate for the identification of individual appliances attending to their consumption. In this way, we will know which electrical appliance is connected to a smart plug, helping to obtain more accurate and efficient load monitoring systems.

The recent increase in smart meters installations in households and small bussiness by electric companies has led to interest in monitoring load techniques in order to provide better quality service and get useful information about appliance usage and user consumption behavior. This works summarizes the current state of the art in Non Intrusive Load Monitoring from its beginning, describes the main process followed in the literature to perform this technique and shows current methods and techniques followed nowadays. The possible application of this techniques in the context of ambient intelligence, energy efficiency, occupancy detection are described. This work also points the current challenges in the field and the future lines of research in this broad topic.

Nowadays, when it comes to achieving goals in business environments or educational environments, the performance successfully has an important role in performing a task. However, this performance can be affected by several factors. One of the most common is the lack of attention. The individual’s attention in performing a task can be determinant for the final quality or even at the task’s conclusion. In this paper is intended to design a solution that can reduce or even eliminate the lack of attention on performing a task. The idea consist on develop an architecture that capture the user behavior through the mouse and keyboard usage. Furthermore, the system will analyze how the devices are used.

The geolocation of people is a field of research that is in continuous development. Currently, the vast majority of systems that can locate people, are used in open spaces that can directly be viewed from the outside using GPS technology. This technology is based on very weak satellite signals and is not suitable for all types of scenarios. When a child gets lost inside a building, the parents inform the guards responsible for building security; an alarm is announced and different actions are undertaken in order to find the child. The steps taken during the search are based on manual procedures; all rooms in the building are searched and security cameras are reviewed. This is why it is crucial to research new solutions which will allow for the localization of individuals in enclosed spaces such as sports stadiums, museums, shopping malls etc. In the last decade, indoor location systems (IPS) have experienced increased growth due to reduced production costs of hardware and miniaturization of device size. This work focuses on the design of an architecture that allows for the monitoring, localization and identification of individuals (especially children or dependent persons) indoors through the use of low cost hardware.

This paper describes a color restoration technique used to remove NIR information from single sensor cameras where color and near-infrared images are simultaneously acquired—referred to in the literature as RGBN images. The proposed approach is based on a neural network architecture that learns the NIR information contained in the RGBN images. The proposed approach is evaluated on real images obtained by using a pair of RGBN cameras. Additionally, qualitative comparisons with a naïve color correction technique based on mean square error minimization are provided.

This paper focuses on near infrared (NIR) image colorization by using a Generative Adversarial Network (GAN) architecture model. The proposed architecture consists of two stages. Firstly, it learns to colorize the given input, resulting in a RGB image. Then, in the second stage, a discriminative model is used to estimate the probability that the generated image came from the training dataset, rather than the image automatically generated. The proposed model starts the learning process from scratch, because our set of images is very different from the dataset used in existing pre-trained models, so transfer learning strategies cannot be used. Infrared image colorization is an important problem when human perception need to be considered, e.g., in remote sensing applications. Experimental results with a large set of real images are provided showing the validity of the proposed approach.

The article describes a strategy to build sentiment lexicons (positive and negative words) from corpora. Special attention will paid to the construction of a domain-specific lexicon from a corpus of movie reviews. Polarity words of the lexicon are assigned weights standing for different degrees of positiveness and negativeness. This lexicon is integrated into a sentiment analysis system in order to evaluate its performance in the task of sentiment classification. The experiments performed shows that the lexicon we generated automatically outperforms other manual lexicons when they are used as features of a supervised sentiment classifier.

One of the main research trends over the last years has focused on knowledge extraction from social networks users. One of the main difficulties of this analysis is the lack of structure of the information and the multiple formats in which it can appear. The present article focuses on the analysis of the information provided by different users in image form. The problem that is intended to be solved is the detection of equal images (although they may have minimal transformations, such as a watermark), which allows establishing links between users who publish the same images. The solution proposed in the article is based on the comparison of hashes, which allows certain transformations that can be made to an image from a computational point of view.

In recent years social networks have increasingly been used to study political opinion formation, monitor electoral campaigns and predict electoral outcomes as they are able to generate huge amount of data, usually in textual and non structured form. In this paper we aim at collecting and analysing data from Twitter posts identifying emerging patterns of topics related to a constitutional referendum that recently took place in Italy to better understand and nowcast its outcome. Using the Twitter API we collect tweets expressing voting intentions in the four weeks before the elections obtaining a database of approximately one million tweets. We restrict the data collection to tweets that contain hashtags referring to the referendum, therefore we are sure to include in the analysis only relevant text. On this huge volume of data, we perform a topic modelling analysis using a Latent Dirichelet Allocation model (LDA) to extract frequent topics and keywords. Analysing the behaviour of frequent words we find that connected to voting in favour of the constitutional reform there are positive words such as future and change while connected to voting against it there are words such ad fear and risk.

Listening to music can affect people emotions. They can experience simultaneous feelings, such as happiness and hope, or sadness and angry, when a song is being played. However, infering emotions that can be caused by a musical fragment is a complex task. To deduce relationships between feelings and music, we propose a sentiment analysis method based on data mining. In particular, different musical features are extracted and classified to analyze the influence of some music parameters on human emotions. In this process, data mining algorithms such as Random k-Labelsets, Multi-Label k-Nearest Neighbors or Apriori have been essential for the success of our proposal.

The interest for providing users with suitable recommendations of songs and playlists has increased since online services for listening to music have become popular. Many methods for achieving this objective have been proposed, some of them addressed to solve well-known problems of recommender systems. However, music application domain has additional drawbacks such as the difficulty for obtaining content information and explicit ratings required by the most reliable recommender methods. In this work, a proposal for improving collaborative filtering methods is presented, whose main advantage is the use of data obtainable easily and automatically from music platforms. The method is based on a procedure for deriving ratings from user implicit behavior as well as on a new way of managing the gray-sheep problem without using content information.

In recent years, with the rise of streaming services like Netflix or Spotify, recommender systems are becoming more and more necessary. The success of Spotify’s Discover Weekly, a music recommender system that suggests new songs to users every week, confirms the need to implement these recommender systems. In this paper we propose a methodology based on collaborative filtering to recommend music for Spotify’s users from an ordered list of the most played songs over a period of time.

This paper describes a proposal to develop a Tourism Recommendation System based in Users and Points-of-Interest (POI) functionality/accessibility levels. The focus is to evaluate if user’s physical and psychological functionality levels can perform an important role in recommendation results accuracy. This work also aims to show the importance of POI classification (accessibility levels are related with each POI ability to receive tourists with certain levels of physical and psychological issues), through the definition of a different model regarding their accessibility and other characteristics.

The objective of this dissertation proposal will focus on finding the computational structures that allow to adapt costly dissimilarity-based classification algorithms to multi-core architectures for CPU based systems, in order to achieve computational efficiency and improving the accelerations of their corresponding sequential implementations. This paper shows preliminary results of the parallel implementation of the leave-one-out test for the Nearest Feature Line and Rectified Nearest Feature Line Segment classifiers.

“Smart Cities” concept is the most predominated in the research world. Its immense benefits to the whole globe like Quality of Life, economic and sustainable development are phenomenal. In order to bring the operational efficiency in each and every axes of Smart City, massive Internet of Things technologies are deploying. In this line all the business houses, municipalities and citizens are striving to invest in Smart Cities. This article is exploring the Indian Smart Cities like Palava, Lavasa, GIFT, and Kochi. The triangulation method is used for evidence, due to contemporary. Study confirmed that Smart City is a collection of Smart (Economy, Environment, Mobility, Living, Governance, and People).

Malware infection in a wireless sensor network (WSN) can represent a potential vulnerability due to the low level of security that these networks exhibit. Consequently, it is very important to study the behavior of the propagation of malware in a WSN. This work aims to design a novel agent-based model to simulate malware spreading. It will provide an efficient software of great help for security administrators.

Advanced persistent threats (APTs) are the new type of cyber attacks that have drastically change the information security landscape. They seek to gather very sensitive information from specific and high-level objectives. The great majority of security tools do not allow handling such an intrusion in a proper way. Consequently, this study aims to analyze its behavior in order to design an agent-based model to simulate the APT cycle of life.

The research proposed here aims at providing a novel organisational metamodel for Large-Scale Multiagent System (LSMASs), with special emphasis on modelling organisational dynamics. Ontology used for the metamodel will comprise selected concepts of human organisation applicable to LSMAS and a particular application domain thereof, Massively Multi-player Online Role-Playing Game (MMORPGs). Practical results include code-generating feature of the modelled system.

5G stringent requirements entail numerous challenges to overcome. Optimizing Network Function Virtualization Infrastructure Point of Presence (NFVI-PoP) placement in a Fog Computing/5G environment is one of these challenges. To solve this problem, this paper proposes an approach for the NFVI-PoP placement problem solution considering 5G mobile network requirements in a Fog Computing (FC) and Software Defined Networks (SDN) ecosystem.

Supporting active and healthy ageing is important both to improve the quality of life of elderly citizens, help them contribute to society as they grow older and to reduce unsustainable pressure on health systems. From social networks to health and fitness, a lot of mobile devices applications (apps) are being developed every day. The variety and availability is such that people start to think that indeed “there’s an app for everything”. Many of these apps address either problems or characteristics that affect older people and that are related with the ageing process (e.g. memory and visual aids apps). They can effectively help people and are under constant evolution. However, the lack of knowledge about these available technological aids can undermine its dissemination and consequently the help that people really receive, especially those who need it the most: older people. In order to tackle the afore mentioned barriers, particularly those related to the profusion of aid apps and the difficulties it creates to the user to select the appropriate ones, we envisage the development of a multi-agent recommender system. The system will combine the results of different algorithms in order to propose the appropriate apps.

The accelerated proliferation of the Internet of Things (IoT) has laid the foundations for the new paradigm of Industry 4.0 and of digital transformations that now arise in organizations. However, these changes have also created challenges related to the management of the large amounts of data; how to process them, store them and convert them into valuable information enabling for effective and efficient decision making.Currently, the research is in its initial stage; we have reviewed literature on multisensor data fusion, which will provide a complete overview of the methodologies, techniques and recent developments in this field. Then, we examine the data fusion model proposed by Bedworth and O’Brien (2000) called the Omnibus Model, since we will be able to use it in the recognition and extraction of unstructured data patterns, such as those coming from IoT sensors. After applying this technique of extracting patterns with less uncertainty and imprecision, we could establish a predictive model oriented at Industry 4.0 for a multi-sensor industrial environment.

We address privacy and decision making in social networks, building a temperament model of users, employing sentiment analysis on written posts and creating PAD models of users through facial images and designing a method that combines this information into a model. We also propose a method for advising the user based on this calculated model.

In humans, interactions happen as a process taught generation to generation, it is our nature given the implicit need of cooperation within society. When talking about systems, reality is rather different, specifically for systems conformed by heterogeneous agents. In this sense, errors in interactions can occur at any time and for many different reasons. This paper proposes a self-healing mechanism based on model checking, taking advantages of the cloud for ambient assisted living systems.

There are new opportunities for research in the field of smart environments that should be explored. The concepts of smart homes and home automation, are currently in growing expansion in the scientific and research point of view, as the market demands for better solutions in this field. Users want that those spaces smartly adapt to their preferences in a transparent way. This paper describes the process of planning, reasoning and modeling of a Smart Environment, using emerging wearable devices on the market (smart watches, fitness trackers, etc.) and newer technologies like NFC, BLE and Wi-Fi Direct. Enabling the user to optimize the efficiency, comfort, and safety at the environments.

Two of the most important characteristic of a Smart Grid are: (a) working optimally, i.e., using the optimal topology and using optimally all available resources in order to minimize the overall planning costs while improve the reliability indexes; (b) the capability to adapt itself to a contingency, for instance, a load increase/decrease, a fault (automatic repair or removal from service the component in an outage situation, etc.). In these cases the reconfiguration of the distribution system must be performed to reroute supplies of energy to sustain power to all customers. This work will propose a new and innovative methodology with two models using a deterministic optimization technique to support power system planning and operation (reconfiguration) decision making in a smart grid context, in order to minimize the overall costs and the same time improve the reliability indexes.

Modern organisations are large complex systems operating in an increasingly dynamic environment, and are tasked to meet their competitive goals by adopting suitable courses of action. The decisions to select effective courses of action call for deep understanding of various aspects of organisation such as its goals, structure, business-as-usual operational processes, and overall business dynamics. The current state-of-practice of decision-making that relies heavily on human experts is often reported as inadequate. This research proposes an actor-based simulation platform as an analysis aid to evaluate the efficacy of decision alternatives with increased precision and rigour in the context of complex dynamic decision making.

The present paper presents an overview of Luis Gomes PhD proposal, focusing on the problem that will be solved and how it will be solved. In the PhD, µGIM system will be developed for microgrid management and microgrid players’ energy management. The proposal also consists in study and analysis of microgrids demand response programs.

This work proposes a organization-based Multi-Agent System that models Local Electricity Market (MASLEM). A bottom-up approach is implemented to manage energy in this work. In this context, agents are able to connect to each other and the power grid to transact electrical energy, and manage their inside electrical energy independently. A Demand Response Program (DRP) based on Indirect Load Control (ILC) method is also used. The performance of our work is evaluated through an Agent Based Modeling (ABM) implementation.

Power systems have been through deep changes, with their operation in the scope of competitive electricity markets (EM) and the increasingly intensive use of renewable energy sources and distributed generation. This requires new business models able to cope with the new opportunities. Virtual Power Players (VPPs) are a new player type which allows aggregating a diversity of players (distribution Generation, storage units, electrical vehicles, and consumers) to participate in the markets and to provide a set of new services promoting generation and consumption efficiency and to improving players’ benefits. A major task of VPPs is the remuneration of generation and of the services (e.g. market operation costs, and energy reserves) as well as charging energy consumption. This PhD research will contribute by developing fair and strategic remuneration and tariff methodologies, able to allow efficient VPP operation and VPP goals accomplishment in EM.

An emerging concept in railway management is to reduce the distance between consecutive trains, actively controlling their separation in order to enforce a continued safe distance. This concept is referred to as “Virtual Coupling” and it needs highly reliable and fast, real-time wireless communication systems. In this respect, TSCH was introduced in the IEEE 802.15.4e amendment to improve reliability of communication. Recent studies addressed the fast joining time problem where different traffic schedules must be merged when different communication domains come together. In this paper, we study the hoping strategy involved in TSCH to derive the probability of successful hopping upon merging.

In recent years, we have been witnessing a real explosion of information, due in large part to the development in Information and Knowledge Technologies (ICTs). As in-formation is the raw material for the discovery of knowledge, there has been a rapid growth, both in the scientific community and in ICT itself, in the approach and study of the phenomenon called Big Data (BD) [1]. The concept of Smart Grids (SG) has emerged as a way of rethinking how to produce and consume energy imposed by economic, political and ecological issues [2]. To become a reality, SGs must be sup-ported by intelligent and autonomous IT systems, to make the right decisions in real time. Knowledge needed for real-time decision-making can only be achieved if SGs are equipped with systems capable of efficiently managing all the information sur-rounding their ecosystem. Multi-Agent systems have been increasingly used from this purpose. This work proposes a system for the management of information in the context of agent based SG to enable the monitoring, in real time, of the events that occur in the ecosystem and to predict upcoming events.

One of the main challenges in power & energy systems is the development of decision support tools which approach the problem as a whole. In this scope, this work contributes to the increase of the interoperability between heterogeneous agent based systems through the use of ontologies, enabling semantic communications.

Electricity markets are not only a new reality but also a constantly evolving sector, due to the high frequency of changes in their rules. Simulation tools combined with Artificial Intelligence techniques, particularly multi-agent simulation, can result in a sophisticated and very useful tool in this context.

Currently, it is possible to find various tools to deal with the unpredictability of electricity markets. However, they mainly focus on spot markets, disfavouring bilateral negotiations. A multi-agent decision support tool is proposed that addresses the identified gap, supporting players in the pre-negotiation and actual negotiation phases.

The penetration of micro-generation brings complex problems to the energy field. In this way, various simulators were designed to give decision support for the stakeholders, however, they intent to solve very specific problems. The proposed tool enables the interoperability between heterogeneous simulators, to simulate more complex problems.

In our research work we plan to develop a Multi-agent based Recommender System to help e-learning systems recommend the more appropriate learning resources to students. In our approach we will explore the multi-agent technology potentialities to build a solution where multiple collaborative and content filtering algorithms, working together, leads to a higher performance solution than that obtained with individual algorithms.

Activity discovery (AD) is the unsupervised process of discovering activities in data produced from streaming sensor networks that are recording the actions of human subjects.

Mild Cognitive Impairment is a previous state to the diagnosis of dementia. This condition implies changes over superior cognitive functions. Some associated factors as age, gender, the loneliness in elderly people, the level of schooling and the inversion of population pyramid are important aspects to take into account. Verbal fluency alteration is an outstanding phenomenon in Mild Cognitive Impairment patients. This alteration goes against vital aspects as the communication, reason related with the isolation in elderly people. Traditional cognitive training is limited because it can not be administered without specialized personnel. However, in the next years the amount of specialist will be insufficient. Computational Cognitive Training is a novel porpoise to decrease the effects of low personnel. Nonetheless, these systems are susceptible to be improved by means of the use of friendly Human Machine Interfaces as the speech recognition, the voice synthesis and the incorporation of cognitive characteristics as autonomy an anticipation. In this work, an improved Computational Cognitive Training Systems is proposed. This tool is designed under the Hybrid systems framework of Artificial Cognitive Systems. The principal aspects of the proposed design are shown. The preliminary results are presented.

Indoor location facilitates the guidance of user in indoor environment in which it is not possible to use GPS. CPS is available in outdoor but in indoor it is necessary to use other technologies to locate users. Some technologies can be based on radiofrequency or other sensors as compass, camera, barometers etc. I this study, we will include a system to calculate the position of users with Bluetooth and accelerometers, in order to calculate the position at home. The system will incorporate artificial intelligence techniques to calculate the variance of signal with Bluetooth in order to calculate the position of a user.

Context aware systems enable interaction between users and their environment in a way that is transparent. Network systems are deployed in these systems in order to collect values from the environment and later retrieve this data which is then used to modify the environment. To improve the interaction between the user and the environment we should create new and simpler interaction mechanisms. In this study we propose interaction through mobile phones, where the system detects and classifies different movements in order to interact with the different elements found in the environment.

Indoor positioning systems have already been introduced in commercial facilities. Since, the signals transmitted from GPS satellites do not penetrate inside buildings, Wifi, Zigbee and Bluetooth are used for indoor position estimation. In this work, we only focus on the use of Bluetooth due to its advantages such as low power consumption, wide signal range and inexpensive. However, the accuracy of positioning is not sufficient in current technologies. Therefore, this paper proposes an indoor positioning method for improving accuracy, using compass and walking speed with an extended Kalman filter. Preliminary experimental results improves accuracy up to 21.2%.

Beauty is a factor which is really difficult to measure or evaluate objectively, since every person can have their own preferences. However, it is common for beauty patterns to tend to be similar in communities, mainly because of different cultural factors. The present study proposes the creation of a prediction system capable of determining the level of beauty from face images in different communities.

During the last decade, vacuuming robots have become very popular, mainly because of their performance. However, due to the sensors that it includes, its size and its price, many researches have appeared applying this kind of robots in different studies. The present study aims to get the robot to follow an object (a ball) inside an enclosure, without colliding with any object.

In the Demand Response (DR) concepts, we witness several barriers that need to be addressed such as, data transferring from promoting entities to demand side. The Open Automated Demand Response (Open ADR) standard specification is a solution for overcoming these barriers. This PhD work proposes a real business model for DR implementation based on Open ADR technology.