Structure learning of graphical models for task-oriented robot grasping

In the collective imaginaries a robot is a human like machine as any androids in science fiction. However the type of robots that you will encounter most frequently are machinery that do work that is too dangerous, boring or onerous. Most of the robots in the world are of this type. They can be found in auto, medical, manufacturing and space industries. Therefore a robot is a system that contains sensors, control systems, manipulators, power supplies and software all working together to perform a task. The development and use of such a system is an active area of research and one of the main problems is the development of interaction skills with the surrounding environment, which include the ability to grasp objects. To perform this task the robot needs to sense the environment and acquire the object informations, physical attributes that may influence a grasp. Humans can solve this grasping problem easily due to their past experiences, that is why many researchers are approaching it from a machine learning perspective finding grasp of an object using information of already known objects. But humans can select the best grasp amongst a vast repertoire not only considering the physical attributes of the object to grasp but even to obtain a certain effect. This is why in our case the study in the area of robot manipulation is focused on grasping and integrating symbolic tasks with data gained through sensors. The learning model is based on Bayesian Network to encode the statistical dependencies between the data collected by the sensors and the symbolic task. This data representation has several advantages. It allows to take into account the uncertainty of the real world, allowing to deal with sensor noise, encodes notion of causality and provides an unified network for learning. Since the network is actually implemented and based on the human expert knowledge, it is very interesting to implement an automated method to learn the structure as in the future more tasks and object features can be introduced and a complex network design based only on human expert knowledge can become unreliable. Since structure learning algorithms presents some weaknesses, the goal of this thesis is to analyze real data used in the network modeled by the human expert, implement a feasible structure learning approach and compare the results with the network designed by the expert in order to possibly enhance it.

Data dissemination on overlay networks through gossip

Gossip protocols have been analyzed as a feasible solution for data dissemination on peer-to-peer networks. In this thesis, a new data dissemination protocol is proposed and compared with other known gossip mechanisms. Performance evaluation is based on simulation.

Microsegmentazione, un approccio scalabile alla sicurezza basato su overlay network

Il crescente numero di attacchi condotti contro sistemi e servizi informatici richiede nuove strategie per la cybersicurezza. In questa tesi si prende in considerazione uno degli approcci più moderni per questa attività, basato su architetture Zero Trust, che deperimetrizzano i sistemi e mirano a verificare ogni tentativo di accesso alle risorse indipendentemente dalla provenienza locale o remota della richiesta. In tale ambito, la tesi propone una nuova forma di microsegmentazione agent-based basata su overlay network, con l'obiettivo di migliorare la scalabilità e la robustezza delle soluzioni esistenti, ad oggi messe in secondo piano in favore della facilità di configurazione. Una consistente serie di test dimostra che l'approccio descritto, attuabile in molteplici tipologie di sistemi cloud, è in grado di garantire, oltre alla sicurezza, scalabilità al crescere dei nodi partecipanti, robustezza evitando punti unici di fallimento e semplicità di configurazione.