books by Andrea Vinci
The growth of data volume collected in urban contexts opens up to their exploitation for improvin... more The growth of data volume collected in urban contexts opens up to their exploitation for improving citizens’ quality-of-life and city management issues, like resource planning (water, electricity), traffic, air and water quality, public policy and public safety services. Moreover, due to the large-scale diffusion of GPS and scanning devices, most of the available data are geo-referenced. Considering such an abundance of data, a very desirable and common task is to identify homogeneous regions in spatial data by partitioning a city into uniform regions based on pollution density, mobility spikes, crimes, or on other characteristics. Density-based clustering algorithms have been shown to be very suitable to detect density-based regions, i.e. areas in which urban events occur with higher density than the remainder of the dataset. Nevertheless, an important issue of such algorithms is that, due to the adoption of global parameters, they fail to identify clusters with varied densities, unless the clusters are clearly separated by sparse regions. In this paper we provide a preliminary analysis about how hierarchical clustering can be used to discover spatial clusters of different densities, in spatial urban data. The algorithm can automatically estimate the area of data having different densities, it can automatically estimate parameters for each cluster so as to reduce the requirement for human intervention or domain knowledge.
Nowadays, the increasing in the use of Internet of Things (IoT) devices is growing the realizatio... more Nowadays, the increasing in the use of Internet of Things (IoT) devices is growing the realization of pervasive Smart Environments (SEs) and Smart Urban Ecosystems, where all the data gathered by the “Things” can be elaborated and used to improve the livability, the safety and the security of the environment, and to make inhabitants lives easier. Many efforts have been already done in the direction of SEs development and in the implementation of platforms specifically designed for SE realization. Anyway, such efforts miss of solutions regarding the interoperability among the realized SEs and other third-part “Things”. This chapter gives an overview of iSapiens, which is a Java-based platform specifically designed for the development and implementation of SEs. iSapiens tries to overcome the interoperability issue by leveraging the Social Internet of Things (SIoT) paradigm that allows to dynamically include in an SE the new “Things” that can appear in an environment without requiring interventions from humans. iSapiens provides tools for the realization of pervasive SEs and relies on the edge computing paradigm. Such paradigm is extremely important in a distributed system since it allows to use distributed storage and computation at the edge of a network, so reducing latencies with respect to move all the executions and storages in the cloud. Moreover, the chapter will review some SE applications realized by exploiting iSapiens concepts.
Smart Cities are smart environments extending over a wide geographical area having the aim of imp... more Smart Cities are smart environments extending over a wide geographical area having the aim of improving the quality of life of the citizens and optimizing the management of city resources. Despite the paramount interest towards these systems, there is a lack of approaches for their design. The Smart Environment Metamodel (SEM) is a framework which is well suited for the development of smart environments in general, and Smart Cities in particular. SEM allows the design of such systems by offering two different perspective focusing on functional and data requirements. This paper aims at showing the effectiveness of SEM by exploiting the framework for the design of a case study referring to a realized Smart City application developed in the city of Cosenza, Italy.
New Internet of Things (IoT) applications are encouraging Smart City and Smart Environments initi... more New Internet of Things (IoT) applications are encouraging Smart City and Smart Environments initiatives all over the world, by leveraging big data and ubiquitous connectivity. This new technology enables systems to monitor, manage and control devices, and to create new knowledge and actionable information, by the real-time analysis of data streams. In order to develop applications in the depicted scenario, the adoption of new paradigms is required. This paper suggests combining the emergent concept of edge/fog computing with the agent metaphor, so as to enable designing systems based on the decentralization of control functions over distributed autonomous and cooperative entities, which run at the edge of the network. Moreover, we suggest the adoption of the iSapiens platform as a reference, as it was designed specifically for the mentioned purposes. Multi-agent applications running on top of iSapiens can create smart services using adaptive and decentralized algorithms which exploit the principles of cognitive IoT.
In the context of time-critical applications there exists the need of clustering data streams so ... more In the context of time-critical applications there exists the need of clustering data streams so as to provide approximated solutions in the shortest possible time, in order to capture in real-time the evolution of physical or social phenomena. In this work, a nature-inspired algorithm for clustering of evolving big data stream is presented, which is designed to be executed on many-core GPU architectures.
New Internet of Things (IoT) applications that leverage ubiquitous connectivity, big data and ana... more New Internet of Things (IoT) applications that leverage ubiquitous connectivity, big data and analytics are enabling Smart City initiatives all over the world. These new applications introduce tremendous new capabilities such as the ability to monitor, manage and control devices remotely, and to create new insights and actionable information from massive streams of real-time data. Supporting this new approach requires the adoption of new paradigms. In this paper, agent tecnology is combined with the emergent concept of Fog computing to design control systems based on the decentralization of control functions over distributed autonomous and cooperative entities that are running at the edge of the network. We describe the Rainbow platform that is designed to bring computation as close as possible to the physical part. Multi-agent systems running on top of Rainbow create smart services using adaptive and decentralized algorithms which exploit the principles of collective intelligence.
Recent advancements in the fields of embedded systems, communication technologies and computer sc... more Recent advancements in the fields of embedded systems, communication technologies and computer science, have laid the foundations for new kinds of applications in which a plethora of physical devices are interconnected and immersed in an environment together with human beings. These so-called Cyber-Physical Systems (CPS) issue a design challenge for new architecture in order to cope with problems such as the heterogeneity of devices, the intrinsically distributed nature of these systems, the lack of reliability in communications, etc. In this paper we introduce Rainbow, an architecture designed to address CPS issues. Rainbow hides heterogeneity by providing a Virtual Object (VO) concept, and addresses the distributed nature of CPS introducing a distributed multi-agent system on top of the physical part. Rainbow aims to get the computation close to the sources of information (i.e., the physical devices) and addresses the dynamic adaptivity requirements of CPS by using Swarm Intelligence algorithms.
Recent advancements in the fields of embedded systems, communication technologies and computer sc... more Recent advancements in the fields of embedded systems, communication technologies and computer science open up to new application scenarios in the home environment. Anyway, many issues raised from the inherent complexity of this new application domain need to be properly tackled. This paper proposes the Cloud-assisted Agent-based Smart home Environment (CASE) architecture for activity recognition with sensors capturing the data related to activities being performed by humans and objects in the environment. Moreover, the potential of analytics methods for discovering activity recognition in such environment has been investigated. CASE easily allows to implement Smart Home applications exploiting a distributed multi-agent system and the cloud technology. The work is mainly focused on activity recognition albeit CASE architecture permits an easy integration of other kinds of smart home applications such as home automation and energy optimization. The CASE effectiveness is shown through the design of a case study consisting of a daily activity recognition of an elder person in its home environment.
This paper focuses on a distributed real time control approach applied to drainage networks. The ... more This paper focuses on a distributed real time control approach applied to drainage networks. The increasing of urbanization and climate change heightens the challenge for new technologies to be developed for drainage networks. Higher runoff volume, produced by the increase in impervious surfaces and intense rain events, overwhelms the existing urban drainage systems. Recent technical improvements have enabled the exploitation of real-time control on drainage networks. The novelty in this paper regards the use of a totally decentralized approach based on a proper combination of a Gossip-based algorithm, which ensures a global correct behaviour even if local faults occur, and a classic controlling technique (PID) used for local actuations.
inproceedings by Andrea Vinci
The management of thermal comfort in a building is a challenging and multi-faced problem because ... more The management of thermal comfort in a building is a challenging and multi-faced problem because it requires considering both objective and subjective parameters that are often in contrast. Subjective parameters are tied to reaching and maintaining an adequate user comfort by considering human preferences and behaviours, while objective parameters can be related to other important aspects like the reduction of energy consumption. This paper exploits cognitive technologies, based on Deep Reinforcement Learning (DRL), for automatically learning how to control the HVAC system in an office. The goal is to develop a cyber-controller able to minimize both the perceived thermal discomfort and the needed energy. The learning process is driven through the definition of a cumulative reward, which includes and combines two reward components that consider, respectively, user comfort and energy consumption. Simulation experiments show that the adopted approach is able to affect the behaviour of the DRL controller and the learning process and therefore to balance the two objectives by weighing the two components of the reward.
Detecting city hotspots in urban environments is a valuable organization methodology for framing ... more Detecting city hotspots in urban environments is a valuable organization methodology for framing detailed knowledge of a metropolitan area, providing high-level summaries for spatial urban datasets. Such knowledge is a valuable support for planner, scientist and policy-maker's decisions. Classic density-based clustering algorithms show to be suitable to discover hotspots characterized by homogeneous density, but their application on multi-density data can produce inaccurate results. For such a reason, since metropolitan cities are heavily characterized by variable densities, multi-density clustering approaches show higher effectiveness to discover city hotspots. Moreover, the growing volumes of data collected in urban environments require high-performance computing solutions, to guarantee efficient, scalable and elastic task executions. This paper describes the design and implementation of a parallel multi-density clustering algorithm, aimed at analyzing high volume of urban data in an efficient way. The experimental evaluation shows that the proposed parallel clustering approach takes out encouraging advantages in terms of execution time and speedup.
Leveraged by a large-scale diffusion of sensing networks and scanning devices in modern cities, h... more Leveraged by a large-scale diffusion of sensing networks and scanning devices in modern cities, huge volumes of geo-referenced urban data are collected every day. Such amount of information is analyzed to discover data-driven models, which can be exploited to tackle the major issues that cities face, including air pollution, virus diffusion, human mobility, traffic flows. In particular, the detection of city hotspots is becoming a valuable organization technique for framing detailed knowledge of a metropolitan area, providing high-level summaries for spatial datasets, which are valuable for planners, scientists, and policymakers. However, while classic density-based clustering algorithms show to be suitable to discover hotspots characterized by homogeneous density, their application on multi-density data can produce inaccurate results. For such a reason, since metropolitan cities are heavily characterized by variable densities, multi-density clustering seems to be more appropriate to discover city hotspots. This paper presents a study about how density-based clustering algorithms are suitable for discovering urban hotspots in a city, by showing a comparative analysis of single-density and multi-density clustering on both state-of-the-art data and real-world data. The experimental evaluation shows that, in an urban scenario, multi-density clustering achieves higher quality hotspots than a single-density approach.
In this work, we propose an IoT edge-based energy management system devoted to minimizing the ene... more In this work, we propose an IoT edge-based energy management system devoted to minimizing the energy cost for the daily-use of in-home appliances. The proposed approach employs a load scheduling based on a load shifting technique, and it is designed to operate in an edge-computing environment naturally. The scheduling considers all together time-variable profiles for energy cost, energy production, and energy consumption for each shiftable appliance. Deadlines for load termination can also be expressed. In order to address these goals, the scheduling problem is formulated as a Markov decision process and then processed through a reinforcement learning technique. The approach is validated by the development of an agent-based real-world test case deployed in an edge context.
The design and implementation of effective systems devoted to the thermal comfort management in a... more The design and implementation of effective systems devoted to the thermal comfort management in a building is a challenging task because they require to consider both objective and subjective parameters, tied for instance to human profile and behavior. This paper presents a novel approach for the management of thermal comfort in buildings by leveraging cognitive technologies, namely the Deep Reinforcement Learning paradigm. The approach is able to learn how to automatically control the HVAC system and improve people's comfort. The learning process is driven by a reward that includes and combines an environmental reward, related to objective environmental parameters, with a human reward, related to subjective human perceptions that are implicitly inferred by the way people interact with the HVAC system. Simulation results aim to assess the impact of the two types of reward on the achieved comfort level.
This paper presents a novel approach for the management of buildings by leveraging cognitive tech... more This paper presents a novel approach for the management of buildings by leveraging cognitive technologies. The proposed approach exploits the Deep Reinforcement Learning paradigm to learn from both a physical and a simulated environment so as to optimize people comfort and energy consumption.
Consolidation of virtual machines (VM) is one of the key strategies used to reduce the power cons... more Consolidation of virtual machines (VM) is one of the key strategies used to reduce the power consumption of Cloud servers. For this reason, it is extensively studied. Consolidation has the goal of allocating virtual machines on a few physical servers as possible while satisfying the Service Level Agreement established with users. Nevertheless, the effectiveness of a con-solidation strategy strongly depends on the forecast of the VMs resource needs. This paper presents the experimental evaluation of a system for energy-aware allocation of virtual machines, driven by predictive data mining models. Migrations are driven by the forecast of the future computational needs (CPU, RAM) of each virtual machine, in order to efficiently allocate those on the available servers. The experimental evaluation, performed on real-world Cloud data traces, reports a comparison of performance achieved by exploiting classification and regression models and shows good benefits in terms of energy saving.
Nowadays, Smart Environments (SEs) are pervasively deployed in buildings (e.g., houses, schools, ... more Nowadays, Smart Environments (SEs) are pervasively deployed in buildings (e.g., houses, schools, and offices) and outdoor environments with the goal of improving the quality of life of their inhabitants. SEs are usually designed and developed by using well-suited architectures and platforms having the aim of simplifying and making straightforward the SE implementation. Up to now, SEs are mostly reactive and, in some ways, proactive. Current research efforts are devoted to making such environments cognitive, i.e., able to automatically adapt and adhere to the possible changes in users' needs and behaviors. Anyway, in this field, the development of SEs is still in its infancy. In this direction, the paper proposes a novel Cognitive-enabled, Edge-based Internet of Things (CEIoT) architecture, purposely designed to develop cognitive IoT-based SEs. Such architecture wants to overcome some limitations arising during the usage of common SE platforms and architectures. CEIoT introduces some abstractions ranging from the "in-platform" implementation of decentralized cognitive algorithms to the realization of smart data aggregations.
The ever increasing diffusion of the Internet of Things is currently promoting the development of... more The ever increasing diffusion of the Internet of Things is currently promoting the development of pervasive Smart Environments. The effectiveness of such systems is highly related to the capability of dealing with possible changes in users' habits, adapting the system to people needs and envisaging people behaviors. For this purposes, it becomes important to have methodological approaches and technologies favoring the development of cognitive systems aware of what is happening inside them. In this paper a methodological approach for the development of context-aware IoT-based Smart Environments is proposed. Such approach relies on a three-layered architecture offering some well suited abstractions taking also into account that computational resources in a system can be located either at the edge of the network or in the Cloud. A case study is proposed which concerns the development of a Smart Office devoted to forecast workers' presence and to adapt the office environmental conditions to them.
Uploads
books by Andrea Vinci
inproceedings by Andrea Vinci