Simulatore di interfaccia uomo-macchina per i controllo di UAV

Recent statistics have demonstrated that two of the most important causes of failures of the UAVs (Uninhabited Aerial Vehicle) missions are related to the low level of decisional autonomy of vehicles and to the man machine interface. Therefore, a relevant issue is to design a display/controls architecture which allows the efficient interaction between the operator and the remote vehicle and to develop a level of automation which allows the vehicle the decision about change in mission. The research presented in this paper focuses on a modular man-machine interface simulator for the UAV control, which simulates UAV missions, developed to experiment solution to this problem. The main components of the simulator are an advanced interface and a block defined automation, which comprehend an algorithm that implements the level of automation of the system. The simulator has been designed and developed following a user-centred design approach in order to take into account the operator’s needs in the communication with the vehicle. The level of automation has been developed following the supervisory control theory which says that the human became a supervisor who sends high level commands, such as part of mission, target, constraints, in then-rule, while the vehicle receives, comprehends and translates such commands into detailed action such as routes or action on the control system. In order to allow the vehicle to calculate and recalculate the safe and efficient route, in term of distance, time and fuel a 3D planning algorithm has been developed. It is based on considering UASs representative of real world systems as objects moving in a virtual environment (terrain, obstacles, and no fly zones) which replicates the airspace. Original obstacle avoidance strategies have been conceived in order to generate mission planes which are consistent with flight rules and with the vehicle performance constraints. The interface is based on a touch screen, used to send high level commands to the vehicle, and a 3D Virtual Display which provides a stereoscopic and augmented visualization of the complex scenario in which the vehicle operates. Furthermore, it is provided with an audio feedback message generator. Simulation tests have been conducted with pilot trainers to evaluate the reliability of the algorithm and the effectiveness and efficiency of the interface in supporting the operator in the supervision of an UAV mission. Results have revealed that the planning algorithm calculate very efficient routes in few seconds, an adequate level of workload is required to command the vehicle and that the 3D based interface provides the operator with a good sense of presence and enhances his awareness of the mission scenario and of the vehicle under his control.

Pervasive Business Intelligence

In the last few years, a new generation of Business Intelligence (BI) tools called BI 2.0 has emerged to meet the new and ambitious requirements of business users. BI 2.0 not only introduces brand new topics, but in some cases it re-examines past challenges according to new perspectives depending on the market changes and needs. In this context, the term pervasive BI has gained increasing interest as an innovative and forward-looking perspective. This thesis investigates three different aspects of pervasive BI: personalization, timeliness, and integration. Personalization refers to the capacity of BI tools to customize the query result according to the user who takes advantage of it, facilitating the fruition of BI information by different type of users (e.g., front-line employees, suppliers, customers, or business partners). In this direction, the thesis proposes a model for On-Line Analytical Process (OLAP) query personalization to reduce the query result to the most relevant information for the specific user. Timeliness refers to the timely provision of business information for decision-making. In this direction, this thesis defines a new Data Warehuose (DW) methodology, Four-Wheel-Drive (4WD), that combines traditional development approaches with agile methods; the aim is to accelerate the project development and reduce the software costs, so as to decrease the number of DW project failures and favour the BI tool penetration even in small and medium companies. Integration refers to the ability of BI tools to allow users to access information anywhere it can be found, by using the device they prefer. To this end, this thesis proposes Business Intelligence Network (BIN), a peer-to-peer data warehousing architecture, where a user can formulate an OLAP query on its own system and retrieve relevant information from both its local system and the DWs of the net, preserving its autonomy and independency.

Il monitoraggio della qualità di vita e dei bisogni di pazienti oncologici e caregiver: il caso ANT

La Fondazione ANT rappresenta una delle più ampie esperienze al mondo di assistenza socio-sanitaria gratuita a domicilio ai malati di tumore, tramite équipe di specialisti costituite da medici, psicologi e infermieri. La patologia oncologica ha un enorme impatto sul benessere dei pazienti. Un modo per raggruppare i diversi sintomi di disagio psicologico è utilizzare il concetto di distress, che sarebbe importante monitorare in modo semplice e veloce. Primo studio: 66 pazienti oncologici (40% uomini; età media 54 anni) in cure palliative domiciliari. Il 79% dei pazienti ha mostrato livelli clinicamente significativi di distress. Il 55% dei partecipanti allo studio ha riportato alti livelli di ansia, e l'81% dei pazienti ha riportato alti livelli di depressione. Dall'analisi delle curve ROC il singolo item del Distress Thermometer, con un cut-off maggiore o uguale a 4, è stato in grado di rilevare il 97% dei soggetti con punteggi clinici di ansia e depressione, quindi può essere utilizzato anche come uno strumento di screening precoce rapido ed affidabile per i disturbi dell'umore. I familiari sono la prima risorsa dei malati di tumore, e l'identificazione dei loro bisogni è utile per individuare chi ha maggiore necessità di aiuto ed in quali aree. Secondo studio: 115 caregiver di pazienti oncologici (37% uomini; età media 52 anni). Di seguito i bisogni più frequenti. Salute psicofisica: “preoccupazioni circa il/la paziente” (72%), ansia (53%) e rabbia (52%). Informazioni: “come prendersi cura del paziente” (64%), “terapie alternative e/o complementari” (64%) e “come gestire lo stress” (57%). Servizi e strutture sanitarie: “un operatore di riferimento”, (65%), “cure infermieristiche a domicilio” (62%), “indicazioni su servizi ospedalieri” (57%), ed “assistenza per caregiver, ad esempio consulenza psicologica” (55%). Il monitoraggio dei bisogni consentirebbe un'ottimizzazione dell'assistenza, prevenendo situazioni che potrebbero compromettere il benessere della famiglia e la qualità dell'assistenza fornita al paziente.

La coazione all'adempimento mediante la tecnica delle misure coercitive indirette

La tesi intende offrire una riflessione in merito all’utilizzo della tecnica delle misure coercitive indirette quali strumenti di coartazione della volontà dell’obbligato al fine di indurlo all’esatto adempimento in presenza di statuizioni giudiziali il cui contenuto si sostanzi in prestazioni a carattere infungibile o in un dovere di astensione dal porre in essere determinati comportamenti, che non possono trovare esecuzione secondo le ordinarie forme previste dal Libro Terzo del Codice di procedura civile. Il lavoro prende le mosse dall'analisi storico-comparatistica dello sviluppo dell'esecuzione processuale indiretta nei principali ordinamenti giuridici europei, per poi soffermarsi sulle singole misure coercitive indirette disciplinate dal legislatore italiano nel corso degli anni al fine di far fronte alle esigenze di tutela emergenti in specifici settori del diritto. Segue un'attenta disamina in ordine alla disposizione di cui all'art. 614-bis c.p.c., con cui è stata introdotta per la prima volta nel nostro ordinamento una misura coercitiva indiretta a carattere generale in materia di esecuzione degli obblighi di fare infungibile e di non fare.

Modeling and optimization techniques for advanced vehicular networking

This thesis deals with optimization techniques and modeling of vehicular networks. Thanks to the models realized with the integer linear programming (ILP) and the heuristic ones, it was possible to study the performances in 5G networks for the vehicular. Thanks to Software-defined networking (SDN) and Network functions virtualization (NFV) paradigms it was possible to study the performances of different classes of service, such as the Ultra Reliable Low Latency Communications (URLLC) class and enhanced Mobile BroadBand (eMBB) class, and how the functional split can have positive effects on network resource management. Two different protection techniques have been studied: Shared Path Protection (SPP) and Dedicated Path Protection (DPP). Thanks to these different protections, it is possible to achieve different network reliability requirements, according to the needs of the end user. Finally, thanks to a simulator developed in Python, it was possible to study the dynamic allocation of resources in a 5G metro network. Through different provisioning algorithms and different dynamic resource management techniques, useful results have been obtained for understanding the needs in the vehicular networks that will exploit 5G. Finally, two models are shown for reconfiguring backup resources when using shared resource protection.

Mapping socialist solidarities: SWAPO, the Namibian liberation struggle, and the Global Cold War

This thesis addresses the entanglements between the Namibian liberation struggle and the global Cold War, focusing on the socialist support provided to the South West African People Organization (SWAPO), the liberation movement that fought for the independence of the country from the South African regime. This thesis aims at analyzing three socialist models of solidarity with the SWAPO’s struggle that developed especially from the late 1970s. Combining archival sources and biographical accounts, it examines the politics of solidarity with SWAPO implemented by East Germany, Cuba, and the Italian Communist Party. The interest lies in understanding how solidarity was declined and received by internal promoters and external addressees. Thus, I explore how these three actors constructed their concept of solidarity with SWAPO according to their national and ideological contexts and how this was received by the SWAPO members who experienced it in various ways. Each socialist actor promoted solidarity with SWAPO by using varying narratives, pursuing their own objectives, and employing diverse instruments, thus carrying out different and sometimes competing visions of socialism and solidarity. On its side, SWAPO was able to take advantage from such visions, as each of them could serve its different needs in diverse ways. In providing a general overview of these three solidarity policies, this thesis has the objective of highlighting the internal pluralization of the “socialist solidarity regime” while at the same time contributing to the debate on the extent of SWAPO’s commitment to socialism during the Namibian liberation struggle. It argues that, while pragmatism has always guided SWAPO during the liberation struggle and the post-independence period, and non-alignment has always been its international stance, socialism has to some extent been a model for the revolution in Namibia, to the point that it is still influencing the SWAPO party today.

How to explain: from theory to practice

Today we live in an age where the internet and artificial intelligence allow us to search for information through impressive amounts of data, opening up revolutionary new ways to make sense of reality and understand our world. However, it is still an area of improvement to exploit the full potential of large amounts of explainable information by distilling it automatically in an intuitive and user-centred explanation. For instance, different people (or artificial agents) may search for and request different types of information in a different order, so it is unlikely that a short explanation can suffice for all needs in the most generic case. Moreover, dumping a large portion of explainable information in a one-size-fits-all representation may also be sub-optimal, as the needed information may be scarce and dispersed across hundreds of pages. The aim of this work is to investigate how to automatically generate (user-centred) explanations from heterogeneous and large collections of data, with a focus on the concept of explanation in a broad sense, as a critical artefact for intelligence, regardless of whether it is human or robotic. Our approach builds on and extends Achinstein’s philosophical theory of explanations, where explaining is an illocutionary (i.e., broad but relevant) act of usefully answering questions. Specifically, we provide the theoretical foundations of Explanatory Artificial Intelligence (YAI), formally defining a user-centred explanatory tool and the space of all possible explanations, or explanatory space, generated by it. We present empirical results in support of our theory, showcasing the implementation of YAI tools and strategies for assessing explainability. To justify and evaluate the proposed theories and models, we considered case studies at the intersection of artificial intelligence and law, particularly European legislation. Our tools helped produce better explanations of software documentation and legal texts for humans and complex regulations for reinforcement learning agents.