No description available.

This research aims to understand the different requirements of civilian multiple autonomous vehicle systems in order to propose an adequate solution for the resource allocation problem. A new classification of unmanned system applications is presented with focus on unmanned aerial vehicles (UAVs). The main resource allocation systems requirements in each category are presented and discussed. A novel dynamic resource allocation model is introduced for efficient sharing of services provided by ad hoc assemblies of heterogeneous autonomous vehicles. A key contribution is the provision of capability to dynamically select sensors and platforms within constraints imposed by time dependencies, refuelling, and transportation services. The resource allocation problem is modelled as a connected network of nodes and formulated as an Integer Linear Program (ILP). Solution fitness is prioritized over computation time. Simulation results of an illustrative scenario are used to demonstrate the ability of the model to plan for sensor selection, refuelling, collaboration and cooperation between heterogeneous resources. Prioritization of operational cost leads to missions that use cheaper resources but take longer to complete. Prioritization of completion time leads to shorter missions at the expense of increased overall resource cost. Missions can be successfully re-planned through dynamic reallocation of new requests during a mission. Monte Carlo studies on systems of increasing complexity show that good solutions can be obtained using low time resolutions, with small time windows at a relatively low computational cost. In comparison with other approaches, the developed ILP model provides provably optimal solutions at the expense of longer computation time. Flight test procedures were developed for performing low cost experiments on a small scale, using commercial off the shelf equipment, with ability to infer conclusions on the large-scale implementation. Flight test experiments were developed and performed that assessed the performance of the developed ILP model and successfully demonstrated its main capabilities.

In recent years, advances in small unmanned aerial vehicle (UAV) technology have transformed the use cases of these aircraft from hobby flying to industrial and business applications. These maneuverable, easily deployed tools can be retrofitted with a myriad of sensors and equipment, which make them suitable to perform a variety of specialized tasks. With increasing UAV capabilities, the function of small UAVs can be extended from pure monitoring or surveillance to the dual objective of monitoring an environment for events and addressing the events in some way. This thesis seeks to explore a subdomain of the dual objective problem described, referred to in this thesis as the multi-UAV persistent search and retrieval task with stochastic target appearance (PSR-STA), in which UAVs continuously search an area over a long period of time for targets of interest, which appear according to a probabilistic model, to retrieve and deliver them to a collector location. The advent of high-speed computers and agent-based modeling theory enable the simulation of multi-UAV PSR-STA. However, it can be complicated to combine parts of multi-UAV PSR-STA such as motion models and multi-UAV coordination into one integrated system, and even after they are combined successfully, it is difficult to analyze the system except with simple comparison tools. This thesis 1) proposes a framework that builds a foundation for understanding how to simulate and analyze multi-UAV PSR-STA through prescribing important design decisions and methods for simulation and 2) identifies metrics, analysis tools, and trends related to overall system effectiveness for multi-UAV PSR-STA. A case study of multi-UAV park cleanup is implemented where many simulations with input parameters chosen by a latin hypercube design of experiments are examined, algorithms for choosing the locations of collectors and charging stations based on probabilistic information are proposed, and the differences in effectiveness between four coverage search patterns are analyzed. Measures are highlighted that provide insight into performance variability over time and space. Line charts and the discrete Fourier transform are used to understand temporal patterns inherent in the data. Principal component analysis is used to analyze relevant spatial patterns in effectiveness, and a random forest surrogate model with a profiler is used to explore the non-linear influence of input parameters on the spatial patterns. The trellis chart or figure of figures method is presented for visualizing spatial and temporal data across many simulations. A second set of experiments based on the park cleanup case study are performed and examined to verify the benefits of these methods.

This work introduces a design and implementation of using multiple unmanned aerial vehicles (UAVs) to achieve cooperative formation flight based on the personal remote sensing platforms developed by the author and the colleagues in the Center for Self-Organizing and Intelligent Systems (CSOIS). The main research objective is to simulate the multiple UAV system, design a multi-agent controller to achieve simulated formation flight with formation reconfiguration and real-time controller tuning functions, implement the control system on actual UAV platforms and demonstrate the control strategy and various formation scenarios in practical flight tests. Research combines analysis on flight control stabilities, develop- ment of a low-cost UAV testbed, mission planning and trajectory tracking, multiple sensor fusion research for UAV attitude estimations, low-cost inertial measurement unit (IMU) evaluation studies, AggieAir remote sensing platform and fail-safe feature development, al- titude controller design for vertical take-off and landing (VTOL) aircraft, and calibration and implementation of an air pressure sensor for wind profiling purposes on the developed multi-UAV platform. Definitions of the research topics and the plans are also addressed.

No description available.

Cette thèse propose des solutions aux problématiques inhérentes au contrôle de formations aériennes de type leader­-suiveur pour des flottes de quadrirotors. Au regard des travaux existants, les stratégies qui sont proposés dans notre travail, considère que le(s) leader{s) a une interaction avec les suiveurs. En outre, les rôles de leader et de suiveur sont interchangeables lors de la formation. Dans un premier temps, la modélisation mathématique d'un seul quadrirotor et celle de la formation de quadrirotors est développée. Ensuite, le problème de suivi de trajectoire pour un seul quadrirotor est étudié. Au travers de l'analyse de 1, dynamique du système pour la conception d'une commande par platitude, il apparait que le suivi de trajectoire pour chaque quadrirotor équivaut à déterminer les sorties plates désirées. Un contrôleur pour système plats permettant l'asservissement des drones pour le suivi de trajectoire est donc proposé. Étant donné la propriété de double-boucle de la dynamique du quadrirotor en boucle fermée, un contrôleur d'attitude avec des grands gains est conçu, selon la théorie « singular perturbation system ». Puisque la dynamique du quadrirotor en boucle fermée fonctionne sur deux échelles de temps, la dynamique de rotation (boundary-layer mode) est contrôlée sur l'échelle de temps la plus rapide. La conception du contrôleur de formation dépend seulement de la dynamique de translation (modèle réduit dans une échelle de temps lente). Ce résultat a simplifié la conception du contrôleur de formation, de telle sorte que le modèle réduit du quadrirotor est utilisé au lieu du modèle complet. Étant donné que le modèle réduit du quadrirotor a une caractéristique de double-intégrateur, un algorithme de consensus pour des systèmes caractérisés par de multiple double-intégrateurs est proposé. Pour traiter le problème de la formation leader-suiveur, une matrice d'interaction est initialement proposée basée sur la matrice de Laplacienne. Nous montrons que la condition de convergence et la vitesse de convergence de l'erreur de formation dépendent de la plus petite valeur propre de la matrice d'interaction. Trois stratégies de contrôle de la formation avec une topologie fixe sont ensuite proposées. Le contrôle de formation par platitude est proposé pour obtenir une formation agressive, tandis que les dérivées de grands ordres de la trajectoire désirée pour chaque UAV sont estimées en utilisant un observateur; la méthode Lyapunov redesign est implémentée pour traiter les non-linéarités de la dynamique de la translation des quadrotors; une loi de commande bornée par l'utilisation, entre autre, de la fonction tangente hyperbolique est développée avec un feedback composite non linéaire, afin d'améliorer les performances de la formation. De plus, une commande de commutation saturée de la formation est étudiée, car la topologie de la formation est variable. La stabilité du système est obtenue grâce aux théories “convex hull » et « common Lyapunov function ». Cette stratégie de commande de commutation permet le changement des leaders dans la formation. Inspirée par certains travaux existants, tels que le contrôle de la formation avec des voisins anonymes, nous proposons, finalement, une loi de commande avec des voisins pondérés, qui montre une meilleure robustesse que le contrôle avec des voisins anonymes. Les résultats de simulation obtenus avec Matlab illustrent premièrement nos stratégies de contrôle que nous proposons De plus, en utilisant le langage de programmation C ++, nos stratégies sont mises en œuvre dans un framework de simulation et d'expérimentation développé au laboratoire Heudiasyc. Grâce aux nombreux tests variés que nous avons réalisés en simulation et en temps-réel, l'efficacité et les avantages de nos stratégies de contrôle de la formation proposées sont présentés. / In this thesis, we address a leader-follower (L-F) formation control problem for multiple UAVs, especially quadrotors. Different from existing works, the strategies, which are proposed in our work, consider that the leader(s) have interaction with the followers. Additionally, the leader(s) are changeable during the formation. First, the mathematical modeling of a single quadrotor and of the formation of quadrotors is developed. The trajectory tracking problem for a single quadrotor is investigated. Through the analysis of the flatness of the quadrotor dynamical model, the desired trajectory for each quadrotor is transferred to the design of the desired at outputs. A flatness-based trajectory tracking controller is, then, proposed. Considering the double-loop property of the closed-loop quadrotor dynamics, a high-gain attitude controller is designed, according to the singular perturbation system theory. Since the closed-loop quadrotor dynamics performs in two time scales, the rotational dynamics (boundary-layer model) is controlled in a fast time scale. The formation controller design is then only considered for the translational dynamics: reduced model in a slow time scale. This result has simplified the formation controller design such that the reduced model of the quadrotor is considered instead of the complete model. Since the reduced model of the quadrotor has a double-integrator characteristic, consensus algorithm for multiple double-integrator systems is proposed. Dealing with the leader-follower formation problem, an interaction matrix is originally proposed based on the Laplacian matrix. We prove that the convergence condition and convergence speed of the formation error are in terms of the smallest eigenvalue of the interaction matrix. Three formation control strategies with fixed formation topology are then proposed. The flatness-based formation control is proposed to deal with the aggressive formation problem, while the high-order derivatives of the desired trajectory for each UAV are estimated by using an observer; the Lyapunov redesign is developed to deal with the nonlinearities of the translational dynamics of the quadrotors; the hyperbolic tangent-based bounded control with composite nonlinear feedback is developed in order to improve the performance of the formation. In an additional way, a saturated switching control of the formation is investigated, where the formation topology is switching. The stability of the system is obtained by introducing the convex hull theory and the common Lyapunov function. This switching control strategy permits the change of the leaders in the formation. Inspired by some existing works, such as the anonymous neighbor-based formation control, we finally propose a weighted neighbor-based control, which shows better robustness than the anonymous neighbor-based control. Simulation results using Matlab primarily illustrate our proposed formation control strategies. Furthermore, using C++ programming, our strategies are implemented on the simulator-experiment framework, developed at Heudiasyc laboratory. Through a variety of tests on the simulator and real-time experiments, the efficiency and the advantages of our proposed formation control strategies are shown. Finally, a vision-based inter-distance detection system is developed. This system is composed by an on-board camera, infrared LEDs and an infrared filter. The idea is to detect the UAVs and calculate the inter-distance by calculating the area of the special LEDs patterns. This algorithm is validated on a PC, with a webcam and primarily implemented on a real quadrotor.

No description available.

The realm of autonomous robotics has seen impressive advancements in recent years, with robots taking on essential roles in various sectors, including disaster response, environmental monitoring, agriculture, and healthcare. As these highly intelligent machines continue to integrate into our daily lives, the pressing imperative is to elevate and refine their performance, enabling them to adeptly manage complex tasks with remarkable efficiency, adaptability, and keen decision-making abilities, all while prioritizing safety-aware navigation, mapping, and control systems. Ensuring the safety-awareness of these robotic systems is of paramount importance in their development and deployment. In this research, bio-inspired neural networks, nature-inspired intelligence, deep learning, heuristic algorithm and optimization techniques are developed for safety-aware autonomous robots navigation, mapping and control. A bio-inspired neural network (BNN) local navigator coupled with dynamic moving windows (DMW) is developed in this research to enhance obstacle avoidance and refines safe trajectories. A hybrid model is proposed to optimize trajectory of the global path of a mobile robot that maintains a safe distance from obstacles using a graph-based search algorithm associated with an improved seagull optimization algorithm (iSOA). A Bat-Pigeon algorithm (BPA) is proposed to undertake adjustable speed navigation of autonomous vehicles in light of object detection for safety-aware vehicle path planning, which can automatically adjust the speed in different road conditions. In order to perform effective collision avoidance in multi-robot task allocation, a spatial dislocation scheme is developed by introduction of an additional dimension for UAVs at different altitudes, whereas UAVs avoid collision at the same altitude using a proposed velocity profile paradigm. A multi-layer robot navigation system is developed to explore row-based environment. A directed coverage path planning (DCPP) fused with an informative planning protocol (IPP) method is proposed to efficiently and safely search the entire workspace. A human-autonomy teaming strategy is proposed to facilitate cooperation between autonomous robots and human expertise for safe navigation to desired areas. Simulation, comparison studies and on-going experimental results of optimization algorithms applied for autonomous robot systems demonstrate their effectiveness, efficiency and robustness of the proposed methodologies.