Η παρούσα διπλωματική εργασία ασχολείται με τον συνεργατικό έλεγχο μιας ομάδας αυτόνομων ρομπότ. Θεωρώντας ότι μόνο ένα ρομπότ έχει πλήρη γνώση για το μονοπάτι που επιθυμούμε να ακολουθήσουν τα ρομπότ, θα πρέπει τα υπόλοιπα παίρνοντας μετρήσεις από διάφορους αισθητήρες και ανταλλάσοντας πληροφορίες μεταξύ τους, να σχεδιάζουν και να ακολουθούν τέτοιες τροχιές ώστε να πετύχουν τον στόχο τους. Από τη στιγμή που ένα μοναδικό ρομπότ γνωρίζει το επιθυμητό μονοπάτι, πρόκειται για ένα leader-follower σχηματισμό όπου κάθε ρομπότ καλείται να ακολουθήσει αυτό που προηγείται. Προκειμένου τα ρομπότ να μπορούν να ακολουθήσουν μία τροχιά αναπτύχθηκε σε περιβάλλον LabVIEW ένας ελεγκτής παρακολούθησης τροχιάς. Για την ανταλλαγή πληροφοριών και μηνυμάτων συντονισμού μεταξύ των ρομπότ, αναπτύχθηκε ασύρματο δίκτυο ZigBee. Το πρόβλημα γνώσης σχετικά με το που βρίσκεται ο leader κάθε ρομπότ λύθηκε με τη χρήση αλγορίθμων που επεξεργάζονται τις μετρήσεις των ενσωματωμένων Sonar. Επίσης το πρόβλημα γνώσης του πραγματικού προσανατολισμού κάθε ρομπότ, αντιμετωπίστηκε με την κατασκευή ενός ψηφιακού μαγνητομέτρου. Για κάθε ρομπότ ξεχωριστά, αναπτύχθηκε ο συνεργατικός αλγόριθμος ο οποίος εξασφαλίζει ότι η ομάδα θα πετυχαίνει τον στόχο που έχει αρχικά τεθεί. Τέλος παρουσιάζονται πειραματικά αποτελέσματα για ομάδες δύο και τριών ρομπότ. / The present thesis elaborates on the cooperative control of mobile robots. Assuming that one robot has complete knowledge of the desired path the robotic platoon must follow, a coordination control scheme must be created, based on sensor measurements and platoon communication, that generates desired trajectories for each of the members of the platoon. From the moment a single robot has complete knowledge of the path, the coordination scheme is based on leader-follower formation control. A trajectory tracking controller was developed in LabVIEW, while a ZigBee based wireless network was implemented for the platoon communication. To find the relative position of the leader for each robot, a sonar based localization algorithm was created, with position measurements through the robot’s encoders and orientation given from a magnetometer. For each robot seperately, the coordination algorithm was developed, that ensures that the platoon will achieve the original goal.

This thesis presents a prototype of Unmanned Cooperative Fire-Seeking and -Fighting Robots (UCSFRs) which have a new way to climb up the stairs or traverse over obstacles with a ball screw. There are three unmanned vehicles (one Mother Vehicle (MV) and two Daughter Vehicles (DVs)) presented in this research. The MV can carry two DVs to climb stairs. They can communicate with each other using Bluetooth communication modules. The core system of the UCSFRs is a PIC 16F877 microcontroller on a 2840 development board. The software is written in C language and the interface is established through Hyper Terminal built in Windows XP. UCSFRs are low cost unmanned vehicles compared with other commercial ones. The double-deck structure is applied on the DVs. The body of the MV can be extended for special purposes. In this research, there are three tests used to verify the functionality of the UCSFRs: (1) MV?s finding and stopping fire, (2) Communication between the MV and the DVs, and (3) the MV?s climbing stairs. In the second test, the DVs run in the opposite direction to assist MV detect fire. By cooperative work, they can save time finding the fire. The MV will go to the hightemperature area according to the data sent by the DVs. Because of the features mentioned above, UCSFRs can be used to perform dangerous tasks instead of fire-fighters.

Made available in DSpace on 2014-12-17T14:56:05Z (GMT). No. of bitstreams: 1 MarceloMY.pdf: 692352 bytes, checksum: bba1f91bbb77a56af37aa6e9e2a536a6 (MD5) Previous issue date: 2005-02-04 / Este trabalho apresenta o desenvolvimento de um m?todo de coordena??o e coopera??o para uma frota de mini-rob?s m?veis. O escopo do desenvolvimento ? o futebol de rob?s. Trata-se de uma plataforma bem estruturada, din?mica e desenvolvida no mundo inteiro. O futebol de rob?s envolve diversos campos do conhecimento incluindo: vis?o computacional, teoria de controle, desenvolvimento de circuitos microcontrolados, planejamento cooperativo, entre outros. A t?tulo de organiza??o os sistema foi dividido em cinco m?dulos: rob?, vis?o, localiza??o, planejamento e controle. O foco do trabalho se limita ao m?dulo de planejamento. Para auxiliar seu desenvolvimento um simulador do sistema foi implementado. O simulador funciona em tempo real e substitui os rob?s reais. Dessa forma os outros m?dulos permanecem praticamente inalterados durante uma simula??o ou execu??o com rob?s reais. Para organizar o comportamento dos rob?s e produzir a coopera??o entre eles foi adotada uma arquitetura hierarquizada: no mais alto n?vel est? a escolha do estilo de jogo do time; logo abaixo decide-se o papel que cada jogador deve assumir; associado ao papel temos uma a??o espec?fica e finalmente calcula-se a refer?ncia de movimento do rob?. O papel de um rob? dita o comportamento do rob? na dada ocasi?o. Os pap?is s?o alocados dinamicamente durante o jogo de forma que um mesmo rob? pode assumir diferentes pap?is no decorrer da partida

No description available.

O problema da tolerância a falhas em robôs manipuladores cooperativos conectados rigidamente a um objeto indeformável é estudado nesta tese. A tolerância a falhas é alcançada através de reconfiguração do sistema de controle. Primeiro, a falha é detectada e isolada. Então, o sistema de controle é reconfigurado de acordo com a falha isolada. As falhas em robôs manipuladores são primeiramente estudadas de acordo com suas consequências no sistema cooperativo. Quatro tipos de falhas são identificados: juntas com balanço livre (sem atuadores ativos), bloqueadas, com informação incorreta de posição e com informação incorreta de velocidade. A detecção e a isolação dos dois primeiros tipos de falhas são alcançadas através de um sistema utilizando redes neurais artificiais. Redes do tipo MLP são empregadas para mapear a dinâmica dos robôs cooperativos sem falhas e uma rede RBF é utilizada para a classificação do vetor de resíduos. As falhas do tipo informação incorreta de posição ou velocidade das juntas são detectadas e isoladas através do uso das restrições impostas pela cadeia cinemática fechada presente no sistema cooperativo. Quando falhas do tipo juntas com balanço livre ou bloqueadas são isoladas, as leis de controle são reconfiguradas. Para estes casos, controladores híbridos de movimento e esmagamento do objeto são deduzidos. Quando falhas do tipo informação incorreta de posição ou velocidade das juntas são isoladas, as medidas afetadas são substituídas por valores estimados. Resultados obtidos em simulações e em robôs cooperativos reais mostram que a metodologia proposta é viável. / The problem of fault tolerance in cooperative manipulators rigidly connected to an undeformable load is addressed in this work. Fault tolerance is reached by reconfiguration of the control system. The faults are firstly detected and isolated. Then, the control system is reconfigured according to the isolated fault. Four faults are considered: free-swinging joint faults, locked joint faults, incorrectly measured joint position faults, and incorrectly measured joint velocity faults. Free-swinging and locked joint faults are detected and isolated by artificial neural networks. MLP’s are utilized to reproduce the dynamics of the fault-free system and an RBF is used to classify the residual vector. Incorrectly measured joint position and velocity faults are detected and isolated based on the kinematic constraints imposed on the cooperative system. When free-swinging and locked joint faults are isolated, the control laws are reconfigured. Control laws for motion and squeeze of the object are developed in these cases. When incorrectly measured joint position faults and incorrectly measured joint velocity faults are isolated, the faulty measurements are replaced by their estimates. Results obtained in simulations and in real cooperative robots indicate that the proposed methodology is viable.

O problema da tolerância a falhas em robôs manipuladores cooperativos conectados rigidamente a um objeto indeformável é estudado nesta tese. A tolerância a falhas é alcançada através de reconfiguração do sistema de controle. Primeiro, a falha é detectada e isolada. Então, o sistema de controle é reconfigurado de acordo com a falha isolada. As falhas em robôs manipuladores são primeiramente estudadas de acordo com suas consequências no sistema cooperativo. Quatro tipos de falhas são identificados: juntas com balanço livre (sem atuadores ativos), bloqueadas, com informação incorreta de posição e com informação incorreta de velocidade. A detecção e a isolação dos dois primeiros tipos de falhas são alcançadas através de um sistema utilizando redes neurais artificiais. Redes do tipo MLP são empregadas para mapear a dinâmica dos robôs cooperativos sem falhas e uma rede RBF é utilizada para a classificação do vetor de resíduos. As falhas do tipo informação incorreta de posição ou velocidade das juntas são detectadas e isoladas através do uso das restrições impostas pela cadeia cinemática fechada presente no sistema cooperativo. Quando falhas do tipo juntas com balanço livre ou bloqueadas são isoladas, as leis de controle são reconfiguradas. Para estes casos, controladores híbridos de movimento e esmagamento do objeto são deduzidos. Quando falhas do tipo informação incorreta de posição ou velocidade das juntas são isoladas, as medidas afetadas são substituídas por valores estimados. Resultados obtidos em simulações e em robôs cooperativos reais mostram que a metodologia proposta é viável. / The problem of fault tolerance in cooperative manipulators rigidly connected to an undeformable load is addressed in this work. Fault tolerance is reached by reconfiguration of the control system. The faults are firstly detected and isolated. Then, the control system is reconfigured according to the isolated fault. Four faults are considered: free-swinging joint faults, locked joint faults, incorrectly measured joint position faults, and incorrectly measured joint velocity faults. Free-swinging and locked joint faults are detected and isolated by artificial neural networks. MLP’s are utilized to reproduce the dynamics of the fault-free system and an RBF is used to classify the residual vector. Incorrectly measured joint position and velocity faults are detected and isolated based on the kinematic constraints imposed on the cooperative system. When free-swinging and locked joint faults are isolated, the control laws are reconfigured. Control laws for motion and squeeze of the object are developed in these cases. When incorrectly measured joint position faults and incorrectly measured joint velocity faults are isolated, the faulty measurements are replaced by their estimates. Results obtained in simulations and in real cooperative robots indicate that the proposed methodology is viable.

Dans un contexte d’Industrie 4.0, on perçoit de nouveaux usages possibles des manipulateurs mobiles (MMs), des robots généralement obtenus par l’association d’un bras manipulateur et d’une plate-forme mobile. Ce travail de thèse se focalise sur la synthèse et la commande de nouveaux MMs coopératifs en définissant trois défis à relever. Le premier défi concerne l’élargissement des domaines d’utilisation des robots par la possibilité de leur utilisation coopérative. Nous définissons ainsi un système robotique modulaire basé sur l’utilisation d’entités robotiques appelés mono-robots (m-bots). Ceux-ci sont des MMs qui peuvent se réarranger sous forme de poly-robot (p-bot) pour réaliser une tâche en collaboration. Le deuxième défi se focalise sur la définition de l’architecture cinématique élémentaire de ces robots. Ainsi, nous proposons une démarche générique de synthèse structurale qui permet l’obtention de plusieurs architectures de m-bots respectant les cahiers des charges relatifs à la tâche en tant que m-bot, mais aussi en tant que p-bot pour un environnement considéré. Cette démarche est basée sur l’analyse structurale des MMs à l’aide des paramètres structuraux des mécanismes (connectivité, mobilité, redondance et hyperstatisme). Le troisième défi proposé est d’arriver à modéliser et contrôler les architectures de MMs synthétisées pour la tâche. Deux lois de commande (PID et hybride force-position) sont proposées pour la réalisation de la tâche considérée. Leur validation a été réalisée grâce à des simulations avancées. / In recent years, the concept of Industry 4.0 has led to new possibilities of use for mobile manipulators (MMs) that are generally made of a manipulator arm mounted on a mobile base. The current Ph.D. is focused on the synthesis and control of new cooperative MMs by defining three challenges. The first challenge concerns the widening of the fields of application of robots. Therefore, we define a modular robotic system based on the use of multiple MMs (mono robots or m-bots) that can be used as a global system (poly-robot or p-bot) for collaborative tasks. The second challenge concerns the definition of the kinematic structure of the MMs. We propose a new generic method of structural synthesis that allows to obtain multiple kinematic architectures for m-bots that respect the constraints imposed by the task and the workspace. This method is based on structural analysis of MMs by the evaluation of the structural parameters (connectivity, mobility, redundancy and overconstraint). The last challenge concerns the modelling and control of the new architectures for the new fields of application. Two control laws (PID control and hybrid force-position control) are proposed in order to realise the considered task. Their validation is done with advanced simulations.