Using multi-agent systems to execute a variety of missions such as environmental monitoring and target tracking has been made possible by the advances in control techniques and computational capabilities. Communication abilities between agents allow them to coact and execute several coordinated missions, among which there is optimal coverage. The optimal coverage problem has several applications in engineering theory and practice, as for example in environmental monitoring, which belongs to the broad class of resource allocation problems, in which a finite number of mobile agents have to be deployed in a given spatial region with the assignment of a sub-region to each agents with respect to a suitable coverage metric. The coverage metric encodes the sensing performance of individual agent with respect to points inside the domain of interest, and a distribution of risk density. Usually the risk density function measures the relative importance assigned to inner regions. The optimal coverage problem in which the risk density is time-invariant has been widely studied in previous research. The solution to this class of problems is centroidal Voronoi tessellation, in which each agent is located on the centroid of the related Voronoi cell. However, there are many scenarios that require to be modelled by time-varying risk density rather than time-invariant one, as for example in area coverage problems where the environment evolves independently of the evolution for the robotic agents deployed to cover the area. In this work, the changing environment is modeled by a time-varying density function which is governed by a convection-diffusion equation. Mixed boundary conditions are considered to model a scenario in which a diffusive substance (e.g., oil from a leaking event or radioactive material from a nuclear accident) enters the area with convective component from the boundary. A non-autonomous feed- back law is employed whose generated trajectories maximize the coverage metric. The asymptotic stability of the multi-agent system is proven by using Barbalat’s lemma, and then theoretical predictions are illustrated by several simulations that represent idealized scenarios.

An area coverage control law in cooperation with reinforcement learning techniques is proposed for deploying multiple autonomous agents in a two-dimensional planar area. A scalar field characterizes the risk density in the area to be covered yielding nonuniform distribution of agents while providing optimal coverage. This problem has traditionally been addressed in the literature to date using locational optimization and gradient descent techniques, as well as proportional and proportional-derivative controllers. In most cases, agents' actuator energy required to drive them in optimal configurations in the workspace is not considered. Here the maximum coverage is achieved with minimum actuator energy required by each agent. Similar to existing coverage control techniques, the proposed algorithm takes into consideration time-varying risk density. These density functions represent the probability of an event occurring (e.g., the presence of an intruding target) at a certain location or point in the workspace indicating where the agents should be located. To this end, a coverage control algorithm using reinforcement learning that moves the team of mobile agents so as to provide optimal coverage given the density functions as they evolve over time is being proposed. Area coverage is modeled using Centroidal Voronoi Tessellation (CVT) governed by agents. Based on [1,2] and [3], the application of Centroidal Voronoi tessellation is extended to a dynamic changing harbour-like environment. The proposed multi-agent area coverage control law in conjunction with reinforcement learning techniques is implemented in a distributed manner whereby the multi-agent team only need to access information from adjacent agents while simultaneously providing dynamic target surveillance for single and multiple targets and feedback control of the environment. This distributed approach describes how automatic flocking behaviour of a team of mobile agents can be achieved by leveraging the geometrical properties of centroidal Voronoi tessellation in area coverage control while enabling multiple targets tracking without the need of consensus between individual agents. Agent deployment using a time-varying density model is being introduced which is a function of the position of some unknown targets in the environment. A nonlinear derivative of the error coverage function is formulated based on the single-integrator agent dynamics. The agent, aware of its local coverage control condition, learns a value function online while leveraging the same from its neighbours. Moreover, a novel computational adaptive optimal control methodology based on work by [4] is proposed that employs the approximate dynamic programming technique online to iteratively solve the algebraic Riccati equation with completely unknown system dynamics as a solution to linear quadratic regulator problem. Furthermore, an online tuning adaptive optimal control algorithm is implemented using an actor-critic neural network recursive least-squares solution framework. The work in this thesis illustrates that reinforcement learning-based techniques can be successfully applied to non-uniform coverage control. Research combining non-uniform coverage control with reinforcement learning techniques is still at an embryonic stage and several limitations exist. Theoretical results are benchmarked and validated with related works in area coverage control through a set of computer simulations where multiple agents are able to deploy themselves, thus paving the way for efficient distributed Voronoi coverage control problems.

This dissertation focuses on real-time decision-making for large-scale domain search and object classification using Multiple Autonomous Vehicles (MAV). In recent years, MAV systems have attracted considerable attention and have been widely utilized. Of particular interest is their application to search and classification under limited sensory capabilities. Since search requires sensor mobility and classification requires a sensor to stay within the vicinity of an object, search and classification are two competing tasks. Therefore, there is a need to develop real-time sensor allocation decision-making strategies to guarantee task accomplishment. These decisions are especially crucial when the domain is much larger than the field-of-view of a sensor, or when the number of objects to be found and classified is much larger than that of available sensors. In this work, the search problem is formulated as a coverage control problem, which aims at collecting enough data at every point within the domain to construct an awareness map. The object classification problem seeks to satisfactorily categorize the property of each found object of interest. The decision-making strategies include both sensor allocation decisions and vehicle motion control. The awareness-, Bayesian-, and risk-based decision-making strategies are developed in sequence. The awareness-based approach is developed under a deterministic framework, while the latter two are developed under a probabilistic framework where uncertainty in sensor measurement is taken into account. The risk-based decision-making strategy also analyzes the effect of measurement cost. It is further extended to an integrated detection and estimation problem with applications in optimal sensor management. Simulation-based studies are performed to confirm the effectiveness of the proposed algorithms.

The objective of this research is to develop a framework for multi-robot coordination and control with emphasis on human-swarm and inter-agent interactions. We focus on two problems: in the first we address how to enable a single human operator to externally influence large teams of robots. By directly imposing density functions on the environment, the user is able to abstract away the size of the swarm and manipulate it as a whole, e.g., to achieve specified geometric configurations, or to maneuver it around. In order to pursue this approach, contributions are made to the problem of coverage of time-varying density functions. In the second problem, we address the characterization of inter-agent interactions and enforcement of desired interaction patterns in a provably safe (i.e., collision free) manner, e.g., for achieving rich motion patterns in a shared space, or for mixing of sensor information. We use elements of the braid group, which allows us to symbolically characterize classes of interaction patterns. We further construct a new specification language that allows us to provide rich, temporally-layered specifications to the multi-robot mixing framework, and present algorithms that significantly reduce the search space of specification-satisfying symbols with exactness guarantees. We also synthesize provably safe controllers that generate and track trajectories to satisfy these symbolic inputs. These controllers allow us to find bounds on the amount of safe interactions that can be achieved in a given bounded domain.

Το κυρίως αντικείμενο της διατριβής αυτής είναι ο σχεδιασμός και η ανάλυση αποκεντρωμένων τεχνικών ελέγχου για επίτευξη μέγιστης κάλυψης από κινούμενα δίκτυα αισθητήρων. Λόγω των πολλών εφαρμογών αυτών σε αποστολές σχετιζόμενες με εξερεύνηση περιοχών ενδιαφέροντος, περιβαλλοντική δειγματοληψία, φύλαξη ή ακόμα και θέματα ασφάλειας, μία μεγάλη μερίδα της επιστημονικής κοινότητας έχει στρέψει το ενδιαφέρον της στην ανάπτυξη μεθόδων για βέλτιστη (ει δυνατόν) περιβαλλοντική αντίληψη μέσω αισθητήρων από αυτόνομες ομάδες ρομποτικών συστημάτων. Τέτοιες ομάδες, συνήθως τοποθετούμενες αρχικώς στις περιοχές ενδιαφέροντος, σχεδιάζονται με στόχο τον αποκεντρωμένο έλεγχό τους, αντί ενός καθολικού εποπτικού συστήματος, με στόχο να επιτύχουν στην εκάστοτε αποστολή. Στα πρώτα στάδια της διατριβής αυτής, το πρόβλημα της κάλυψης μιας περιοχής ενδιαφέροντος από μία ομάδα όμοιων κόμβων αναλύεται από υπολογιστική σκοπιά. Οι κινούμενοι κόμβοι υποθέτονται ότι υπακούν σε απλοϊκό κινηματικό μοντέλο διακριτού χρόνου, ενώ η αισθητήρια επίδοσή τους θεωρείται ακτινική, περιορισμένης εμβέλειας, ομοιόμορφη γύρω από τον κόμβο. Σαν πρώτη προσέγγιση, η κατεύθυνση σε κάθε χρονική στιγμή για βέλτιστη κάλυψη καθορίζεται βάσει τεχνικών διαμέρισης του χώρου βασιζόμενες στην έννοια της απόστασης. Η αναπτυσσόμενη στρατηγική επιτρέπει σταδιακή αύξηση της καλυπτόμενης επιφάνειας μεταξύ διαδοχικών βημάτων, ενώ έχει ως απαίτηση την κίνηση ενός μόνο επιτρεπτού κόμβου τη φορά. Στη συνέχεια, το προαναφερθέν σχέδιο επεκτείνεται για την περίπτωση ετερογενών δικτύων, όπου η ετερογένεια αντικατοπτρίζεται στις άνισες εμβέλειες απόδοσης αίσθησης των κόμβων. Επιπροσθέτως, επέκταση σε μοντέλο συνεχούς χρόνου επιτρέπει την κίνηση όλων των κόμβων του δικτύου ταυτόχρονα, αυξάνοντας ιδιαίτερα τον χρόνο σύγκλισης προς την βέλτιστη κατάσταση, ειδικά για μεγάλης κλίμακας δίκτυα. Μία εναλλακτική διαμέριση του χώρου αναπτύσσεται, η οποία βασίζεται κυρίως στα αισθητήρια μοτίβα των κόμβων, παρά στις θέσεις των κόμβων καθεαυτές. Τα παραγόμενα κελιά του χώρου ανατιθέμενα στους κόμβους αποτελούν τον βασικό πυρήνα του αλγόριθμου οργάνωσης, με στόχο την αποκεντρωμένη οργάνωση της κινούμενης ομάδας, ώστε να επιτύχει βέλτιστη απόδοση κάλυψης. Υποκινούμενοι από την υψηλού–βαθμού ανισοτροπία που χαρακτηρίζει κάποιους τύπους αισθητήρων, όπως κατευθυντικά μικρόφωνα για ανίχνευση ήχου σε εφαρμογές ασφάλειας, ή ακόμα μοτίβα εκπομπής/λήψης κατευθυντικών κεραιών σε σενάρια τηλεπικοινωνιακής κάλυψης, η έρευνά μας επεκτείνεται πέραν του κλασσικού ακτινικού μοντέλου δίσκου αίσθησης. Βασιζόμενοι σε συγκεκριμένες ιδιότητες για επίπεδες κυρτές καμπύλες, μια αποκεντρωμένη στρατηγική οργάνωσης αναπτύχθηκε για δίκτυα που χαρακτηρίζονται από κυρτά αισθητήρια μοτίβα ίδιας κατευθυντικότητας. Παρότι η κυρτότητα των συνόλων αίσθησης φαίνεται να θέτει ένα μεγάλου βαθμού περιορισμό στο συνολικό πρόβλημα, στην πραγματικότητα προσπερνάται μέσω ανάθεσης αυτών ως το μέγιστο κυρτό χωρίο που εγγράφεται στο πρωταρχικώς ανισοτροπικό μοτίβο. Το σχήμα ελέγχου επεκτείνεται στη συνέχεια για την περίπτωση όπου εισάγουμε ένα επιπλέον βαθμό ελευθερίας στις κινηματικές ικανότητες των κόμβων, ενσωματώνοντας έτσι διαφορετικές και χρονικά μεταβαλλόμενες κατευθυντικότητες μεταξύ των μοτίβων αυτών. Το παραγόμενο πλάνο ελέγχου αποδεικνύεται ότι οδηγεί ανισοτροπικά δίκτυα σε βέλτιστες τοπολογίες, αναφορικά με τα αισθητήρια μοτίβα τους, ελέγχοντας κατάλληλα ταυτόχρονα την θέση και προσανατολισμό, μέσω ενός καινοτόμου σχήματος κατακερματισμού του χώρου βασιζόμενο στο εκάστοτε μοτίβο. Η διατριβή κλείνει με την μελέτη δικτύων με περιορισμούς στην εμβέλεια επικοινωνίας αναφορικά με την μετάδοση πληροφοριών μεταξύ των κόμβων. Στην πλειονότητα των σχετικών εργασιών, το ζήτημα αυτό προσπερνάται επιτρέποντας στην εμβέλεια επικοινωνίας να είναι τουλάχιστον διπλάσια αυτής της (ομοιόμορφης) αίσθησης, εγγυώντας έτσι την αποκεντρωμένη φύση των πλάνων ελέγχου. Ο προτεινόμενος έλεγχος επιτρέπει την αποσύζευξη μεταξύ των δύο αυτών εμβελειών, οδηγώντας το δίκτυο στην βέλτιστη κατάσταση, μέσω ταυτόχρονου σεβασμού του εκάστοτε, εκ των προτέρων δοσμένου, περιορισμού στην εμβέλεια επικοινωνίας. Συγκεντρωτικά συμπεράσματα και συγκριτική ανάλυση παρουσιάζονται στο τελευταίο κεφάλαιο, ενώ προτείνονται μελλοντικά πλάνα επέκτασης των τεχνικών αυτών. / The main scope of this thesis is the design and analysis of distributed control strategies for achieving optimum area coverage in mobile sensor networks. Due to the numerous applications of the latter in missions as area exploration, environmental sampling, patrolling, or even security, a large part of the scientific community has turned its interest on developing methods for achieving optimum, if possible, sensing environmental perception by groups of autonomous mobile agents. Such robotic teams, randomly deployed in areas of interest initially, are designed to coordinate their motion in a distributed manner, rather than via a global supervisory system, in order to succeed in the corresponding mission objective. At the first stages of this thesis, the coverage problem of an area of interest by a group of identical nodes is examined from a numerical point of view. The mobile nodes are considered to be governed by simple discrete–time kinodynamic motion, while their sensing performance is assumed radial, range–limited, uniform around the node. As a first approach, the optimum direction at each time step for optimum deployment achievement is determined based on proper distance–based space partitioning techniques. The developed concept allows for gradual increase in the covered area among consecutive steps, although suffers from allowing motion of one node at a time. In the sequel, the aforementioned concept is extended to the case of heterogeneous networks, where heterogeneity lays mainly in the unequal limited–range of the sensing performance of the nodes. In addition, extension to continuous–time allows for simultaneous motion of the nodes, increasing drastically the convergence time towards the optimal state, especially for large–scale networks. An alternate partitioning of the space is developed that is mainly based on the nodes’ footprints, rather than their spatial positions only. The resulting assigned cells form the main core for the coordination algorithm proposed, in order to distributedly organize the mobile swarm to achieve optimum sensing performance. Motivated by the high–degree anisotropy that governs the sensing domains of certain types of sensors, i.e. directional microphones for sound sensing mainly for security applications, or even the radiation patterns of directional antennas in communication–coverage scenarios, our research is extended beyond the standard disc model of sensing. Based on certain properties for planar convex curves, a distributed strategy is developed for networks characterized by convex sensing domains of same orientation. Although convexity of the sensing sets may seem to impose a high level restriction to the overall setup, in fact can be assigned as the maximal convex inscribed set in any (originally) anisotropic pattern. The control scheme is further extended, in the sequel, for the case of adding an extra degree of freedom to the node’s mobility abilities, incorporating different and time–varying orientations among the nodes patterns. The resulting scheme is proven to lead anisotropic networks in optimum configurations, considering their sensing footprints, by properly controlling both the nodes’ positions and orientations, via an innovative pattern–based partitioning scheme of the sensed space. The thesis ends by examining the case where radio–range constraints are imposed on inter–agents communication. In the majority of the related works, this issues is usually overcome by allowing RF range as double the sensing one, guaranteeing that way distributed nature of the control schemes. The proposed scheme allows for uncorrelated RF and sensing ranges in the network, while guarantees convergence of the network towards the optimal state, via simultaneous preservation of a–priori imposed radio–range constraints. Concluding remarks along with comparative discussion are presented in the last chapter, where future research plans and ways to improve the already developed schemes are proposed.

<div>Multi-robot systems have demonstrated strong capability in handling environmental operations. In this study, We examine how a team of robots can be utilized in covering and removing spill patches in a dynamic environment by executing three consecutive stages: deployment, coverage, and rendezvous. </div><div> </div><div>For the deployment problem, we aim for robot allocation based on the discreteness of the patches that need to be covered. With the deep neural network (DNN) based spill detector and remote sensing facilities such as drones with vision sensors and satellites, we are able to obtain the spill distribution in the workspace. Then, we formulate the allocation problem in a general optimization form and provide solutions using an integer linear programming (ILP) solver under several realistic constraints. After the allocation process is completed and the robot team is divided according to the number of spills, we deploy robots to their computed optimal goal positions. In the robot deployment part, control laws based on artificial potential field (APF) method are proposed and practiced on robots with a common unicycle model. </div><div> </div><div>For the coverage control problem, we show two strategies that are tailored for a wirelessly networked robot team. We propose strategies for coverage with and without path planning, depending on the availability of global information. Specifically, in terms of coverage with path planning, we partition the workspace from the aerial image into pieces and let each robot take care of one of the pieces. However, path-planning-based coverage relies on GPS signals or other external positioning systems, which are not applicable for indoor or GPS-denied circumstances. Therefore, we propose an asymptotic boundary shrink control that enables a collective coverage operation with the robot team. Such a strategy does not require a planned path, and because of its distributedness, it shows many advantages, including system scalability, dynamic spill adaptability, and collision avoidance. In case of a large-scale patch that poses challenges to robot connectivity maintenance during the operation, we propose a pivot-robot coverage strategy by mean of an a priori geometric tessellation (GT). In the pivot-robot-based coverage strategy, a team of robots is sent to perform complete coverage to every packing area of GT in sequence. Ultimately, the entire spill in the workspace can be covered and removed.</div><div> </div><div>For the rendezvous problem, we investigate the use of graph theory and propose control strategies based on network topology to motivate robots to meet at a designated or the optimal location. The rendezvous control strategies show a strong robustness to some common failures, such as mobility failure and communication failure. To expedite the rendezvous process and enable herding control in a distributed way, we propose a multi-robot multi-point rendezvous control strategy. </div><div> </div><div>To verify the validity of the proposed strategies, we carry out simulations in the Robotarium MATLAB platform, which is an open source swarm robotics experiment testbed, and conduct real experiments involving multiple mobile robots.</div>

Proteins are considered the primary actors in living organisms. Proteins mainly perform their functions by interacting with other proteins. Protein-protein interactions underpin various biological activities such as metabolic cycles, signal transduction, and immune response. PPI identification has been addressed by various experimental methods such as the yeast two-hybrid, mass spectrometry, and protein microarrays, to mention a few. However, due to the sheer number of proteins, experimental methods for finding interacting and non-interacting protein pairs are time-consuming and costly. Therefore a sequence-based framework called ProtInteract is developed to predict protein-protein interaction. ProtInteract comprises two components: first, a novel autoencoder architecture that encodes each protein's primary structure to a lower-dimensional vector while preserving its underlying sequential pattern by extracting uncorrelated attributes and more expressive descriptors. This leads to faster training of the second network, a deep convolutional neural network (CNN) that receives encoded proteins and predicts their interaction. Three different scenarios formulate the prediction task. In each scenario, the deep CNN predicts the class of a given encoded protein pair. Each class indicates different ranges of confidence scores corresponding to the probability of whether a predicted interaction occurs or not. The proposed framework features significantly low computational complexity and relatively fast response. The present study makes two significant contributions to the field of protein-protein interaction (PPI) prediction. Firstly, it addresses the computational challenges posed by the high dimensionality of protein datasets through the use of dimensionality reduction techniques, which extract highly informative sequence attributes. Secondly, the proposed framework, ProtInteract, utilises this information to identify the interaction characteristics of a protein based on its amino acid configuration. ProtInteract encodes the protein's primary structure into a lower-dimensional vector space, thereby reducing the computational complexity of PPI prediction. Our results provide evidence of the proposed framework's accuracy and efficiency in predicting protein-protein interactions.