Dissertationen zum Thema „Apprentissage robotique“
Geben Sie eine Quelle nach APA, MLA, Chicago, Harvard und anderen Zitierweisen an
Machen Sie sich mit Top-50 Dissertationen für die Forschung zum Thema "Apprentissage robotique" bekannt.
Neben jedem Werk im Literaturverzeichnis ist die Option "Zur Bibliographie hinzufügen" verfügbar. Nutzen Sie sie, wird Ihre bibliographische Angabe des gewählten Werkes nach der nötigen Zitierweise (APA, MLA, Harvard, Chicago, Vancouver usw.) automatisch gestaltet.
Sie können auch den vollen Text der wissenschaftlichen Publikation im PDF-Format herunterladen und eine Online-Annotation der Arbeit lesen, wenn die relevanten Parameter in den Metadaten verfügbar sind.
Sehen Sie die Dissertationen für verschiedene Spezialgebieten durch und erstellen Sie Ihre Bibliographie auf korrekte Weise.
Filliat, David. „Navigation, perception et apprentissage pour la robotique“. Habilitation à diriger des recherches, Université Pierre et Marie Curie - Paris VI, 2011. http://tel.archives-ouvertes.fr/tel-00649692.
Der volle Inhalt der QuelleDroniou, Alain. „Apprentissage de représentations et robotique développementale : quelques apports de l'apprentissage profond pour la robotique autonome“. Thesis, Paris 6, 2015. http://www.theses.fr/2015PA066056/document.
Der volle Inhalt der QuelleThis thesis studies the use of deep neural networks to learn high level representations from raw inputs on robots, based on the "manifold hypothesis"
Droniou, Alain. „Apprentissage de représentations et robotique développementale : quelques apports de l'apprentissage profond pour la robotique autonome“. Electronic Thesis or Diss., Paris 6, 2015. http://www.theses.fr/2015PA066056.
Der volle Inhalt der QuelleThis thesis studies the use of deep neural networks to learn high level representations from raw inputs on robots, based on the "manifold hypothesis"
Rolland, de Rengervé Antoine. „Apprentissage Interactif en Robotique Autonome : vers de nouveaux types d'IHM“. Phd thesis, Université de Cergy Pontoise, 2013. http://tel.archives-ouvertes.fr/tel-00969519.
Der volle Inhalt der QuelleRolland, de Rengerve Antoine. „Apprentissage Interactif en Robotique Autonome : vers de nouveaux types d'IHM“. Thesis, Cergy-Pontoise, 2013. http://www.theses.fr/2013CERG0664/document.
Der volle Inhalt der QuelleAn autonomous robot collaborating with humans should be able to learn how to navigate and manipulate objects in the same task. In a classical approach, independent functional modules are considered to manage the different aspects of the task (navigation, arm control,...) . To the contrary, the goal of this thesis is to show that learning tasks of different kinds can be tackled by learning sensorimotor attractors from a few task nonspecific structures. We thus proposed an architecture which can learn and encode attractors to perform navigation tasks as well as arm control.We started by considering a model inspired from place-cells for navigation of autonomous robots. On-line and interactive learning of place-action couples can let attraction basins emerge, allowing an autonomous robot to follow a trajectory. The robot behavior can be corrected and guided by interacting with it. The successive corrections and their sensorimotor coding enables to define the attraction basin of the trajectory. My first contribution was to adapt this principle of sensorimotor attractor building for the impedance control of a robot arm. While a proprioceptive posture is maintained, the arm movements can be corrected by modifying on-line the motor command expressed as muscular activations. The resulting motor attractors are simple associations between the proprioceptive information of the arm and these motor commands. I then showed that the robot could learn visuomotor attractors by combining the proprioceptive and visual information with the motor attractors. The visuomotor control corresponds to a homeostatic system trying to maintain an equilibrium between the two kinds of information. In the case of ambiguous visual information, the robot may perceive an external stimulus (e.g. a human hand) as its own hand. According to the principle of homeostasis, the robot will act to reduce the incoherence between this external information and its proprioceptive information. It then displays a behavior of immediately observed gestures imitation. This mechanism of homeostasis, completed by a memory of the observed sequences and action inhibition capability during the observation phase, enables a robot to perform deferred imitation and learn by observation. In the case of more complex tasks, we also showed that learning transitions can be the basis for learning sequences of gestures, like in the case of cognitive map learning in navigation. The use of motivational contexts then enables to choose between different learned sequences.We then addressed the issue of integrating in the same architecture behaviors involving visuomotor navigation and robotic arm control to grab objects. The difficulty is to be able to synchronize the different actions so the robot act coherently. Erroneous behaviors of the robot are detected by evaluating the actions predicted by the model with respect to corrections forced by the human teacher. These situations can be learned as multimodal contexts modulating the action selection process in order to adapt the behavior so the robot reproduces the desired task.Finally, we will present the perspectives of this work in terms of sensorimotor control, for both navigation and robotic arm control, and its link to human robot interface issues. We will also insist on the fact that different kinds of imitation behavior can result from the emergent properties of a sensorimotor control architecture
Aklil, Nassim. „Apprentissage actif sous contrainte de budget en robotique et en neurosciences computationnelles. Localisation robotique et modélisation comportementale en environnement non stationnaire“. Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066225/document.
Der volle Inhalt der QuelleDecision-making is a highly researched field in science, be it in neuroscience to understand the processes underlying animal decision-making, or in robotics to model efficient and rapid decision-making processes in real environments. In neuroscience, this problem is resolved online with sequential decision-making models based on reinforcement learning. In robotics, the primary objective is efficiency, in order to be deployed in real environments. However, in robotics what can be called the budget and which concerns the limitations inherent to the hardware, such as computation times, limited actions available to the robot or the lifetime of the robot battery, are often not taken into account at the present time. We propose in this thesis to introduce the notion of budget as an explicit constraint in the robotic learning processes applied to a localization task by implementing a model based on work developed in statistical learning that processes data under explicit constraints, limiting the input of data or imposing a more explicit time constraint. In order to discuss an online functioning of this type of budgeted learning algorithms, we also discuss some possible inspirations that could be taken on the side of computational neuroscience. In this context, the alternation between information retrieval for location and the decision to move for a robot may be indirectly linked to the notion of exploration-exploitation compromise. We present our contribution to the modeling of this compromise in animals in a non-stationary task involving different levels of uncertainty, and we make the link with the methods of multi-armed bandits
Do, Huu Nicolas. „Apprentissage de représentations sensori-motrices pour la reconnaissance d'objet en robotique“. Phd thesis, Université Paul Sabatier - Toulouse III, 2007. http://tel.archives-ouvertes.fr/tel-00283073.
Der volle Inhalt der QuelleDo, Huu Nicolas Chatila Raja. „Apprentissage de représentations sensori-motrices pour la reconnaissance d'objet en robotique“. Toulouse (Université Paul Sabatier, Toulouse 3), 2008. http://thesesups.ups-tlse.fr/190.
Der volle Inhalt der QuelleSalaün, Camille. „Apprentissage De Modèles Pour La Commande De La Mobilité Interne En Robotique“. Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00545534.
Der volle Inhalt der QuelleLazaric, Nathalie. „Apprentissage organisationnel et développement technologique : la création robotique dans l'industrie automobile allemande“. Compiègne, 1993. http://www.theses.fr/1992COMP563E.
Der volle Inhalt der QuelleBoutin, Luc. „Biomimétisme, génération de trajectoires pour la robotique humanoïde à partir de mouvements humains“. Poitiers, 2009. http://theses.edel.univ-poitiers.fr/theses/2009/Boutin-Luc/2009-Boutin-Luc-These.pdf.
Der volle Inhalt der QuelleThe true reproduction of human locomotion is a topical issue on humanoid robots. The goal of this work is to define a process to imitate the human motion with humanoid robots. In the first part, the motion capture techniques are presented. The measurement protocol adopted is exposed and the calculation of joint angles. An adaptation of three existing algorithms is proposed to detect the contact events during complex movements. The method is valided by measurements on thirty healthy subjects. The second part deals with the generation of humanoid trajectories imitating the human motion. Once the problem and the imitation process are defined, the balance criterion of walking robots is presented. Using data from human motion capture, the reference trajectories of the feet and ZMP are defined. These paths are modified to avoid collision between feet, particularly in the case of executing a slalom. Finally an inverse kinematics algorithm developed for this problem is used to determine the joint angles associated with the robot reference trajectories of the feet and ZMP. Several applications on robots HOAP-3 and HRP-2 are presented. The trajectories are validated according to the robot balance through dynamic simulations of the computed motion, and respecting the limits of actuators
Paquier, Williams. „Apprentissage ouvert de représentations et de fonctionalités en robotique : analyse, modèles et implémentation“. Toulouse 3, 2004. http://www.theses.fr/2004TOU30233.
Der volle Inhalt der QuelleAutonomous acquisition of representations and functionalities by a machine address several theoretical questions. Today’s autonomous robots are developed around a set of functionalities. Their representations of the world are deduced from the analysis and modeling of a given problem, and are initially given by the developers. This limits the learning capabilities of robots. In this thesis, we propose an approach and a system able to build open-ended representation and functionalities. This system learns through its experimentations of the environment and aims to augment a value function. Its objective consists in acting to reactivate the representations it has already learnt to connote positively. An analysis of the generalization capabilities to produce appropriate actions enable define a minimal set of properties needed by such a system. The open-ended representation system is composed of a network of homogeneous processing units and is based on position coding. The meaning of a processing unit depends on its position in the global network. This representation system presents similarities with the principle of numeration by position. A representation is given by a set of active units. This system is implemented in a suite of software called NeuSter, which is able to simulate million unit networks with billions of connections on heterogeneous clusters of POSIX machines. .
Cuperlier, Nicolas. „Apprentissage et prédiction de séquences sensori-motrices : architecture neuromimétique pour la navigation et la planification d'un robot mobile“. Cergy-Pontoise, 2006. http://www.theses.fr/2006CERG0316.
Der volle Inhalt der QuelleNavigation of an autonomous mobile robot in an unknown environment is a complex task that raises numerous issues in perception, categorisation, planning, and motor control. Solving all these problems in an integrated manner remains a challenge for roboticians. Thus, we propose a unified neuronal framework, based on the modeling of different parts of the mammalian brain’s functionalities: the hippocampus, the prefrontal cortex and the basal ganglia. Key topics are the multi-modal data integration like vision (the prevailing input), path integration, motivation, and also the inner and outer interactions between the structures. A first part of our work consists in modeling neural networks able to learn and predict sensory-motor combinations (transition cells) which are inputs of a cognitive map used to plan according to conflicting motivations. The cognitive map is learned without using any Cartesian coordinates nor occupancy grids. Already known transitions are used in exploration in order to preferentially explore unknown zones to reduce exploration time and enhance the completion of the cognitive map. Links of this map are learned or reinforced according to the behavior and enable to take into account dynamical changes of the environment. Exploration periods may be alternated with planning periods. The second part of this thesis brings an interesting solution for computing and selecting the final movement to perform. It also gives a stable motor control. Instead of using a (( Winner Takes All )) mechanism to select the movement, we increase the planned movement accuracy via a soft competition. Hence several movements are proposed and fed in another layer where the final motor command is obtained as the stable solution of a dynamical system: a one dimensional neural field coding for the heading direction. This field allows to endow the system with a final movement selection leading to a better movement generalization and consequently to a more reliable movement while planning. Our model gives a control architecture allowing to exhibit on a mobile robot navigation behaviors inspired from biology. This architecture can be considered as an attempt to explain underlying mechanisms implemented by mammals for these kind of behaviors. Furthermore, we can list the following benefits of our model: on-line localization, active exploration, planning and mapping in an uncompletely explored environment. These benifits cast an original light on the S. L. A. M problem (Simultaneous Localization and Map building of an unknown environment)
Djian, David. „Contribution à l'analyse de scènes par vision active : utilisation de réseaux Bayesiens“. ENSMP, 1997. http://www.theses.fr/1997ENMP0739.
Der volle Inhalt der QuelleIn this thesis, scene analysis is aimed at improving autonomy in robotics. In order to perceive a complex environment, we have chosen vision which provides rich information. The framework of active vision consists in processing only relevant data (regions of interest in an image). Robustness to sensor errors is improved by taking into account uncertainty in the sensor models used. Finally, we study the learning of object models for recognition. We propose a hierarchical model of the environment which incorporates explicitely sensor actions in a graph representing an object. Graph nodes correspond to object parts, and graph links correspond to the perception actions which detect object parts during recognition. Our sensor model is an extension of Henderson's logicial sensors. We have introduced the concept of dynamic observers as a region of interest in the image coupled with elementary vision algorithms and their associated model of uncertainty. Observers are moved in the image thanks to an internal mechanism decided in the sensor models. The recognition process is driven by a strategy of perception which decides where and when dynamic observers should be started. Bayes nets offer a rigorous and unified framework based o probability theory to implement the three components of our recognition system. They can represent prior knowledge about objet models, incremental knowledge gathered during recognition, and various strategies of perception (reasoning under certainty). We show sucessful results for the learning and the recognition of complex object models on real images
Infantes, Guillaume. „Apprentissage de modèles de comportement pour le contrôle d'exécution et la planification robotique“. Phd thesis, Université Paul Sabatier - Toulouse III, 2006. http://tel.archives-ouvertes.fr/tel-00129505.
Der volle Inhalt der QuellePAQUIER, Williams. „Apprentissage ouvert de representations et de fonctionnalites en robotique : anayse, modeles et implementation“. Phd thesis, Université Paul Sabatier - Toulouse III, 2004. http://tel.archives-ouvertes.fr/tel-00009324.
Der volle Inhalt der QuelleHugues, Louis. „Apprentissage de comportements pour un robot autonome“. Paris 6, 2002. http://www.theses.fr/2002PA066414.
Der volle Inhalt der QuelleRenaudo, Erwan. „Des comportements flexibles aux comportements habituels : meta-apprentissage neuro-inspiré pour la robotique autonome“. Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066508/document.
Der volle Inhalt der QuelleIn this work, we study how the notion of behavioral habit, inspired from the study of biology, can benefit to robots. Robot control architectures allow the robot to be able to plan to reach long term goals while staying reactive to events happening in the environment (Kortenkamp et Simmons, 2008). However, these architectures are rarely provided with learning capabilities that would allow them to acquire knowledge from experience. On the other hand, learning has been shown as an essential abiilty for behavioral adaptation in mammals. It permits flexible adaptation to new contexts but also efficient behavior in known contexts (Dickinson, 1985). The learning mechanisms are modeled as model-based (planning) and model-free (habitual) reinforcement learning algorithms (Sutton et Barto, 1998) which are combined into a global model of behavior (Daw et al., 2005). We proposed a robotic control architecture that take inspiration from this model of behavior and embed the two kinds of algorithms, and studied its performance in a robotic simulated task. None of the several methods for combining the algorithm we studied gave satisfying results, however, it allowed to identify some properties required for the planning process in a robotic task. We extended our study to two other tasks (one being on a real robot) and confirmed that combining the algorithms improves learning of the robot's behavior
Renaudo, Erwan. „Des comportements flexibles aux comportements habituels : meta-apprentissage neuro-inspiré pour la robotique autonome“. Electronic Thesis or Diss., Paris 6, 2016. http://www.theses.fr/2016PA066508.
Der volle Inhalt der QuelleIn this work, we study how the notion of behavioral habit, inspired from the study of biology, can benefit to robots. Robot control architectures allow the robot to be able to plan to reach long term goals while staying reactive to events happening in the environment (Kortenkamp et Simmons, 2008). However, these architectures are rarely provided with learning capabilities that would allow them to acquire knowledge from experience. On the other hand, learning has been shown as an essential abiilty for behavioral adaptation in mammals. It permits flexible adaptation to new contexts but also efficient behavior in known contexts (Dickinson, 1985). The learning mechanisms are modeled as model-based (planning) and model-free (habitual) reinforcement learning algorithms (Sutton et Barto, 1998) which are combined into a global model of behavior (Daw et al., 2005). We proposed a robotic control architecture that take inspiration from this model of behavior and embed the two kinds of algorithms, and studied its performance in a robotic simulated task. None of the several methods for combining the algorithm we studied gave satisfying results, however, it allowed to identify some properties required for the planning process in a robotic task. We extended our study to two other tasks (one being on a real robot) and confirmed that combining the algorithms improves learning of the robot's behavior
Barate, Renaud. „Apprentissage de fonctions visuelles pour un robot mobile par programmation génétique“. Phd thesis, Université Pierre et Marie Curie - Paris VI, 2008. http://pastel.archives-ouvertes.fr/pastel-00004864.
Der volle Inhalt der QuelleGrizou, Jonathan. „Apprentissage simultané d'une tâche nouvelle et de l'interprétation de signaux sociaux d'un humain en robotique“. Thesis, Bordeaux, 2014. http://www.theses.fr/2014BORD0146/document.
Der volle Inhalt der QuelleThis thesis investigates how a machine can be taught a new task from unlabeled humaninstructions, which is without knowing beforehand how to associate the human communicative signals withtheir meanings. The theoretical and empirical work presented in this thesis provides means to createcalibration free interactive systems, which allow humans to interact with machines, from scratch, using theirown preferred teaching signals. It therefore removes the need for an expert to tune the system for eachspecific user, which constitutes an important step towards flexible personalized teaching interfaces, a key forthe future of personal robotics.Our approach assumes the robot has access to a limited set of task hypotheses, which include the task theuser wants to solve. Our method consists of generating interpretation hypotheses of the teaching signalswith respect to each hypothetic task. By building a set of hypothetic interpretation, i.e. a set of signallabelpairs for each task, the task the user wants to solve is the one that explains better the history of interaction.We consider different scenarios, including a pick and place robotics experiment with speech as the modalityof interaction, and a navigation task in a brain computer interaction scenario. In these scenarios, a teacherinstructs a robot to perform a new task using initially unclassified signals, whose associated meaning can bea feedback (correct/incorrect) or a guidance (go left, right, up, ...). Our results show that a) it is possible tolearn the meaning of unlabeled and noisy teaching signals, as well as a new task at the same time, and b) itis possible to reuse the acquired knowledge about the teaching signals for learning new tasks faster. Wefurther introduce a planning strategy that exploits uncertainty from the task and the signals' meanings toallow more efficient learning sessions. We present a study where several real human subjects controlsuccessfully a virtual device using their brain and without relying on a calibration phase. Our system identifies, from scratch, the target intended by the user as well as the decoder of brain signals.Based on this work, but from another perspective, we introduce a new experimental setup to study howhumans behave in asymmetric collaborative tasks. In this setup, two humans have to collaborate to solve atask but the channels of communication they can use are constrained and force them to invent and agree ona shared interaction protocol in order to solve the task. These constraints allow analyzing how acommunication protocol is progressively established through the interplay and history of individual actions
Guerry, Joris. „Reconnaissance visuelle robuste par réseaux de neurones dans des scénarios d'exploration robotique. Détecte-moi si tu peux !“ Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLX080/document.
Der volle Inhalt der QuelleThe main objective of this thesis is visual recognition for a mobile robot in difficult conditions. We are particularly interested in neural networks which present today the best performances in computer vision. We studied the concept of method selection for the classification of 2D images by using a neural network selector to choose the best available classifier given the observed situation. This strategy works when data can be easily partitioned with respect to available classifiers, which is the case when complementary modalities are used. We have therefore used RGB-D data (2.5D) in particular applied to people detection. We propose a combination of independent neural network detectors specific to each modality (color & depth map) based on the same architecture (Faster RCNN). We share intermediate results of the detectors to allow them to complement and improve overall performance in difficult situations (luminosity loss or acquisition noise of the depth map). We are establishing new state of the art scores in the field and propose a more complex and richer data set to the community (ONERA.ROOM). Finally, we made use of the 3D information contained in the RGB-D images through a multi-view method. We have defined a strategy for generating 2D virtual views that are consistent with the 3D structure. For a semantic segmentation task, this approach artificially increases the training data for each RGB-D image and accumulates different predictions during the test. We obtain new reference results on the SUNRGBD and NYUDv2 datasets. All these works allowed us to handle in an original way 2D, 2.5D and 3D robotic data with neural networks. Whether for classification, detection and semantic segmentation, we not only validated our approaches on difficult data sets, but also brought the state of the art to a new level of performance
Lyubova, Natalia. „Developmental approach of perception for a humanoid robot“. Palaiseau, École nationale supérieure de techniques avancées, 2013. http://www.theses.fr/2013ESTA0003.
Der volle Inhalt der QuelleGaudiello, Ilaria. „Learning robotics, with robotics, by robotics : a study on three paradigms of educational robotics, under the issues of robot representation, robot acceptance, and robot impact on learning“. Thesis, Paris 8, 2015. http://www.theses.fr/2015PA080081.
Der volle Inhalt der QuelleThrough a psychological perspective, the thesis concerns the three ER learning paradigms that are distinguished upon the different hardware, software, and correspondent modes of interaction allowed by the robot. Learning robotics was investigated under the issue of robot representation. By robot representation, we mean its ontological and pedagogical status and how such status change when users learn robotics. In order to answer this question, we carried out an experimental study based on pre- and post-inquiries, involving 79 participants. Learning with robotics was investigated under the issue of robot’s functional and social acceptance. Here, the underlying research questions were as follows: do students trust in robot’s functional and social savvy? Is trust in functional savvy a pre-requisite for trust in social savvy? Which individuals and contextual factors are more likely to influence this trust? In order to answer these questions, we have carried an experimental study with 56 participants and an iCub robot. Trust in the robot has been considered as a main indicator of acceptance in situations of perceptual and socio-cognitive uncertainty and was measured by participants’ conformation to answers given by iCub. Learning by robotics was investigated under the issue of robot’s impact on learning. The research questions were the following: to what extent the combined RBI & IBSE frame has a positive impact on cognitive, affective, social and meta-cognitive dimensions of learning? Does this combined educational frame improve both domain-specific and non-domain specific knowledge and competences of students? In order to answer these questions, we have carried a one-year RBI & IBSE experimental study in the frame of RObeeZ, a research made through the FP7 EU project Pri-Sci-Net. The longitudinal experiments involved 26 pupils and 2 teachers from a suburb parisian primary school
Chapelle, Jérôme. „Une architecture multi-agents pour un apprentissage autonome guidé par les émotions“. Montpellier 2, 2006. http://www.theses.fr/2006MON20181.
Der volle Inhalt der QuelleLucidarme, Philippe. „Apprentissage et adaptation pour des ensembles de robots réactifs coopérants“. Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2003. http://tel.archives-ouvertes.fr/tel-00641563.
Der volle Inhalt der QuelleHenaff, Patrick. „Commande bio-inspirée et genèse de mouvements rythmiques en robotique“. Habilitation à diriger des recherches, Université de Cergy Pontoise, 2011. http://tel.archives-ouvertes.fr/tel-00667651.
Der volle Inhalt der QuelleDromnelle, Rémi. „Architecture cognitive générique pour la coordination de stratégies d'apprentissage en robotique“. Electronic Thesis or Diss., Sorbonne université, 2021. http://www.theses.fr/2021SORUS039.
Der volle Inhalt der QuelleThe main objective of this thesis is to propose a new method for online adaptation of robotic learning, allowing robots to dynamically and autonomously adapt their behavior according to variations in their own performance. The developed method is sufficiently general and task-independent that a robot using it can perform different dynamic tasks of various nature without any algorithm or parameter adjustment by the programmer. The algorithms underlying this method consist of a meta-control system that allows the robot to call upon two decision-making experts following a different behavioral strategy. The model-based expert builds a model of the effects of long-term actions and uses this model to decide; this strategy is computationally expensive, but quickly converges to the solution. The model-free expert is inexpensive in terms of computational resources, but takes time to converge to the optimal solution. In this work, we have developed a new criterion for the coordination of these two experts allowing the robot to dynamically change its strategy over time. We show in this work that our behavior coordination method allows the robot to maintain an optimal performance in terms of performance and computation time. We also show that the method can cope with abrupt changes in the environment, changes in goals or changes in the behavior of the human partner in the case of interaction tasks
Lesort, Timothée. „Continual Learning : Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes“. Thesis, Institut polytechnique de Paris, 2020. http://www.theses.fr/2020IPPAE003.
Der volle Inhalt der QuelleHumans learn all their life long. They accumulate knowledge from a sequence of learning experiences and remember the essential concepts without forgetting what they have learned previously. Artificial neural networks struggle to learn similarly. They often rely on data rigorously preprocessed to learn solutions to specific problems such as classification or regression.In particular, they forget their past learning experiences if trained on new ones.Therefore, artificial neural networks are often inept to deal with real-lifesuch as an autonomous-robot that have to learn on-line to adapt to new situations and overcome new problems without forgetting its past learning-experiences.Continual learning (CL) is a branch of machine learning addressing this type of problems.Continual algorithms are designed to accumulate and improve knowledge in a curriculum of learning-experiences without forgetting.In this thesis, we propose to explore continual algorithms with replay processes.Replay processes gather together rehearsal methods and generative replay methods.Generative Replay consists of regenerating past learning experiences with a generative model to remember them. Rehearsal consists of saving a core-set of samples from past learning experiences to rehearse them later. The replay processes make possible a compromise between optimizing the current learning objective and the past ones enabling learning without forgetting in sequences of tasks settings.We show that they are very promising methods for continual learning. Notably, they enable the re-evaluation of past data with new knowledge and the confrontation of data from different learning-experiences. We demonstrate their ability to learn continually through unsupervised learning, supervised learning and reinforcement learning tasks
Spach, Michel. „Activités robotiques à l'école primaire et apprentissage de concepts informatiques : quelle place du scénario pédagogique ? Les limites du co-apprentissage“. Thesis, Sorbonne Paris Cité, 2017. http://www.theses.fr/2017USPCB198/document.
Der volle Inhalt der QuelleThis research, which takes place within the framework of Baron and Bruillard's research in didactics of computer science,analyzes how primary school teachers, not computer experts, design and implement scenarios involving ground pedagogical robots in their classrooms. The integration of these robots has been studied with the aim of shedding light on their possible pedagogical contributions. It shows how these teachers succeed in defining pedagogical situations of these knowledge objects to which they have never been confronted before and in developing pupils' thinking in action. Student activity was analyzed, through the instrumental approach (Rabardel), in order to understand how the learning of computational concepts emerges from these activities. The question of the learning of concepts and methods specific to the computer domain through robotics is analyzed using the theory of conceptual fields (Vergnaud). This research provides additional understanding how these teachers intuitively develop and implement scenarios to teach a few computer concepts. It demonstrates their ability to integrate tangible or symbolic objects into computer learning sessions by performing a minimum analysis of the robot's functionality. During the activities in which they are mobilized, robotic tools and teaching aids accompany learning. In terms of learning, pupils have forged, through instrumentation and instrumentalities, tools and methods to understand the computer object. The concepts and notions involved are particularly dependent on the technological contexts specific to each robot. Methods specific to software production allowed the sequencing of the programming activity into phases of specification, design, realization and development. Programming paradigms were also approached, such as procedural programming in the case of the Bee-Bot robot and event programming in the case of the study of the behavior of the Thymio robot. Outside the computer field, problem solving, by being placed at the heart of the scenarios, allowed students to develop trial and error approaches in a small group work environment that facilitate exchanges and interactions between students
Marquez-Gamez, David. „Vers une navigation visuelle en environnement dynamique inconnu : apprentissage et exécution de trajectoire avec détection et suivi d'objets mobiles“. Phd thesis, INSA de Toulouse, 2012. http://tel.archives-ouvertes.fr/tel-00842378.
Der volle Inhalt der QuelleBeaussé, Nils. „Apprentissage visuo-moteur, implication pour le développement sensorimoteur et l’émergence d'interactions sociales“. Thesis, Cergy-Pontoise, 2019. http://www.theses.fr/2019CERG1051.
Der volle Inhalt der QuelleThis thesis try to bring answers to the question of the sensorimotor learning and development in the context of human-robot interactions in a real non-constrained environment. To achieve this goal we defend in this thesis the fact that human being interacts through intentional and conscious strategy but also depends of the property of their low level motor system, their body, and of their sensorimotor learning loops, allowing these to facilitate implicitly this interaction. We try to answer these questions through the study of the sensorimotor loops in humans, and through the study of the development of these properties in infants. First, we study here the properties of our robot « Tino », which is a prototype of an humanoid hydraulic robot, unique in France and which is the main experimental platform used in this thesis. We analyses in this thesis the property of this robot and made analogies with the human motor system properties that are implied in the interaction between human, the environment and other humans. We show how certain of these properties could be used to simplify tasks for the control system. We study finally the limit of this analogy and of the exploitation of these properties. After this part we study in this thesis the modeling of low level motor loop and of the properties of the human muscular system in order to capture the main interesting properties for interactions. We propose an implementation on robot and analyses the properties of this control system in simulation and on the robotic platform Tino. Then, we propose a bio-inspired and developmental neural architecture that is able to learn visuomotor association with babbling exploration of the environment. We show with this model implemented on the robot Tino that we can observe the emergence of implicit social interaction through the sensorimotor loops, such are imitation and pointing gesture. But this model use a simple associative learning which is able to construct an “actions repertoire” but is unable to react to the environment and humans finely. To solve this problem we have developed, through two simulations, a learning model based on reinforcement learning to allow our system to produce coherent trajectory in order to act in an environment. We applied this in a simulated task of grasping and moving an object on a table. We show then the analogies between this model and historic experiments about the impact of intention on the motor actions and trajectories in humans Finally we study in this thesis the dynamic of interactions and the interest of bringing oscillatory neural network in these sensorimotor architectures. To this end we propose in this thesis several oscillatory models able to learn and to adapt in the context of bio-inspired architecture that learn in interaction with a real environment
Martinez, Margarit Aleix. „Apprentissage visuel dans un système de vision active : application dans un contexte de robotique et reconnaissance du visage“. Paris 8, 1998. http://www.theses.fr/1998PA081521.
Der volle Inhalt der QuelleOuanezar, Sofiane. „Contrôle moteur par le cervelet et interface Cerveau-Machine pour commander un doigt robotique“. Phd thesis, Télécom ParisTech, 2010. http://pastel.archives-ouvertes.fr/pastel-00577959.
Der volle Inhalt der QuelleRaiola, Gennaro. „Co-manipulation with a library of virtual guides“. Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLY001/document.
Der volle Inhalt der QuelleRobots have a fundamental role in industrial manufacturing. They not only increase the efficiency and the quality of production lines, but also drastically decrease the work load carried out by humans.However, due to the limitations of industrial robots in terms of flexibility, perception and safety, their use is limited to well-known structured environment. Moreover, it is not always cost-effective to use industrial autonomous robots in small factories with low production volumes.This means that human workers are still needed in many assembly lines to carry out specific tasks.Therefore, in recent years, a big impulse has been given to human-robot co-manipulation.By allowing humans and robots to work together, it is possible to combine the advantages of both; abstract task understanding and robust perception typical of human beings with the accuracy and the strength of industrial robots.One successful method to facilitate human-robot co-manipulation, is the Virtual Guides approach which constrains the motion of the robot along only certain task-relevant trajectories. The so realized virtual guide acts as a passive tool that improves the performances of the user in terms of task time, mental workload and errors.The innovative aspect of our work is to present a library of virtual guides that allows the user to easily select, generate and modify the guides through an intuitive haptic interaction with the robot.We demonstrated in two industrial tasks that these innovations provide a novel and intuitive interface for joint human-robot completion of tasks
Cogrel, Benjamin, und Benjamin Cogrel. „Sélection contextuelle de services continus pour la robotique ambiante“. Phd thesis, Université Paris-Est, 2013. http://tel.archives-ouvertes.fr/tel-00961567.
Der volle Inhalt der QuelleBideaux, Eric. „Stan : systeme de transport a apprentissage neuronal. application de la vision omnidirectionnelle a la localisation d'un robot mobile autonome“. Besançon, 1995. http://www.theses.fr/1995BESA2008.
Der volle Inhalt der QuelleMoualla, Aliaa. „Un robot au Musée : Apprentissage cognitif et conduite esthétique“. Thesis, CY Cergy Paris Université, 2020. http://www.theses.fr/2020CYUN1002.
Der volle Inhalt der QuelleIn my thesis I treat the subject of autonomous learning based on social referencing in a real environment, "the museum". I am interested in adding and analyzing the mechanisms necessary for a robot to pursue such a type of learning. I am also interested in the impact of a specific and individual learning to each robot on the whole of a group of robots confronted with a known situation or on the contrary new, more precisely:In the first chapter, we will discuss in a didactic way the tools needed to understand the models and methods that we will use throughout our work. We will discuss the basics of neural formalism, conditioning learning, categorization, and dynamic neural fields.In the second chapter, we will briefly present the biological visual system then we will review a state of the art of different models dealing with visual perception and object recognition. As part of a bio-inspired approach, we will then present the model of the visual system of the "Berenson" robot, the sensorimotor architecture allowing to associate an emotional value with an observed object. Then we study the performances of the visual system with and without space competition mechanism.In the third chapter we will move to the level of human-machine interactions, we will show that the interest of visitors to the robot does not only depend on its shape, but on its behavior and more specifically its ability to interact on an emotional level. (here facial expressions). We first analyze the impact of the visual system on the low level control of robot actions. We show that the low level of the spatial competition between the values associated with the zones of interest of the image is important for the recognition of objects and thus affects the coherence of the behavior of the robot and therefore the legibility of this behavior. . We then introduce modifications on the control of eye, head and body movements inspired by biological processes (change of the frame of reference). In the end, we analyze the tests performed in the museum to assess the readability of the behavior of the robot (its movements and facial expressions).In the fourth chapter, our work continues with the addition of inspired bio-based neural mechanisms that allow the emergence of important joint attention capacity to achieve more "natural" interactions with visitors to the museum but also to discuss a point from a theoretical point of view the emergence of the notion of agency. Berenson represents today a form of experimentation unique in the social sciences as in development robotics.In the fifth chapter, we will focus on evaluating the effect of the emergence of aesthetic preferences on a whole population of robots (in simulation). We argue that the variability of learning offered by special environments such as a museum leads to the individuation of robots. We also question the interest of teaching artificial systems using a single large database in order to improve their performance. Avoiding a uniform response to an unknown situation in a population of individuals increases its chances of success
Chenu, Alexandre. „Leveraging sequentiality in Robot Learning : Application of the Divide & Conquer paradigm to Neuro-Evolution and Deep Reinforcement Learning“. Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS342.
Der volle Inhalt der Quelle“To succeed, planning alone is insufficient. One must improvise as well.” This quote from Isaac Asimov, founding father of robotics and author of the Three Laws of Robotics, emphasizes the importance of being able to adapt and think on one’s feet to achieve success. Although robots can nowadays resolve highly complex tasks, they still need to gain those crucial adaptability skills to be deployed on a larger scale. Robot Learning uses learning algorithms to tackle this lack of adaptability and to enable robots to solve complex tasks autonomously. Two types of learning algorithms are particularly suitable for robots to learn controllers autonomously: Deep Reinforcement Learning and Neuro-Evolution. However, both classes of algorithms often cannot solve Hard Exploration Problems, that is problems with a long horizon and a sparse reward signal, unless they are guided in their learning process. One can consider different approaches to tackle those problems. An option is to search for a diversity of behaviors rather than a specific one. The idea is that among this diversity, some behaviors will be able to solve the task. We call these algorithms Diversity Search algorithms. A second option consists in guiding the learning process using demonstrations provided by an expert. This is called Learning from Demonstration. However, searching for diverse behaviors or learning from demonstration can be inefficient in some contexts. Indeed, finding diverse behaviors can be tedious if the environment is complex. On the other hand, learning from demonstration can be very difficult if only one demonstration is available. This thesis attempts to improve the effectiveness of Diversity Search and Learning from Demonstration when applied to Hard Exploration Problems. To do so, we assume that complex robotics behaviors can be decomposed into reaching simpler sub-goals. Based on this sequential bias, we try to improve the sample efficiency of Diversity Search and Learning from Demonstration algorithms by adopting Divide & Conquer strategies, which are well-known for their efficiency when the problem is composable. Throughout the thesis, we propose two main strategies. First, after identifying some limitations of Diversity Search algorithms based on Neuro-Evolution, we propose Novelty Search Skill Chaining. This algorithm combines Diversity Search with Skill- Chaining to efficiently navigate maze environments that are difficult to explore for state-of-the-art Diversity Search. In a second set of contributions, we propose the Divide & Conquer Imitation Learning algorithms. The key intuition behind those methods is to decompose the complex task of learning from a single demonstration into several simpler goal-reaching sub-tasks. DCIL-II, the most advanced variant, can learn walking behaviors for under-actuated humanoid robots with unprecedented efficiency. Beyond underlining the effectiveness of the Divide & Conquer paradigm in Robot Learning, this work also highlights the difficulties that can arise when composing behaviors, even in elementary environments. One will inevitably have to address these difficulties before applying these algorithms directly to real robots. It may be necessary for the success of the next generations of robots, as outlined by Asimov
Fontbonne, Nicolas. „Individual and group learning dynamics in evolutionary collective robotics“. Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS069.
Der volle Inhalt der QuelleWith their proliferation in industry and daily life, robots are now increasingly required to interact with each other. This thesis deals with the problem of coordination between robots in a context where they have to learn their control policy autonomously. These policies are optimized with machine learning algorithms that take advantage of a reward function to increase performance incrementally. The structure of this function will significantly influence the learning dynamics and, then, the possible behaviours of the agents. We first study systems where agents individually receive a local reward adapted to their actions and must converge towards an optimal collective behaviour. We introduce a distributed evolutionary learning algorithm called Horizontal Information Transfert (HIT) that tackles this particular issue. Agents interact on-line in their environment and must learn their control policy with an embedded evolutionary algorithm and a parameter exchange system. It has the advantage of coping with the limited computation and communication capabilities of low-cost robots, which are often used in swarm robotics. We analyze this algorithm's characteristics and learning dynamics on a foraging task. We then study systems where the reward is given globally to the entire team. Therefore, this evaluation does not necessarily represent each agent's performance, and it can be challenging to calculate an individual contribution. We introduce a centralized cooperative co-evolutionary algorithm (CCEA) that modulates the number of agents' policies modification to find a compromise between evaluation quality and execution speed. This modulation also helps in completing tasks where improving team performance requires multiple agents to update in a synchronized manner. We use a multi-robot resource selection problem and a simulated multi-rover exploration problem to provide experimental validations of the proposed algorithms
Risser-Maroix, Olivier. „Similarité visuelle et apprentissage de représentations“. Electronic Thesis or Diss., Université Paris Cité, 2022. http://www.theses.fr/2022UNIP7327.
Der volle Inhalt der QuelleThe objective of this CIFRE thesis is to develop an image search engine, based on computer vision, to assist customs officers. Indeed, we observe, paradoxically, an increase in security threats (terrorism, trafficking, etc.) coupled with a decrease in the number of customs officers. The images of cargoes acquired by X-ray scanners already allow the inspection of a load without requiring the opening and complete search of a controlled load. By automatically proposing similar images, such a search engine would help the customs officer in his decision making when faced with infrequent or suspicious visual signatures of products. Thanks to the development of modern artificial intelligence (AI) techniques, our era is undergoing great changes: AI is transforming all sectors of the economy. Some see this advent of "robotization" as the dehumanization of the workforce, or even its replacement. However, reducing the use of AI to the simple search for productivity gains would be reductive. In reality, AI could allow to increase the work capacity of humans and not to compete with them in order to replace them. It is in this context, the birth of Augmented Intelligence, that this thesis takes place. This manuscript devoted to the question of visual similarity is divided into two parts. Two practical cases where the collaboration between Man and AI is beneficial are proposed. In the first part, the problem of learning representations for the retrieval of similar images is still under investigation. After implementing a first system similar to those proposed by the state of the art, one of the main limitations is pointed out: the semantic bias. Indeed, the main contemporary methods use image datasets coupled with semantic labels only. The literature considers that two images are similar if they share the same label. This vision of the notion of similarity, however fundamental in AI, is reductive. It will therefore be questioned in the light of work in cognitive psychology in order to propose an improvement: the taking into account of visual similarity. This new definition allows a better synergy between the customs officer and the machine. This work is the subject of scientific publications and a patent. In the second part, after having identified the key components allowing to improve the performances of thepreviously proposed system, an approach mixing empirical and theoretical research is proposed. This secondcase, augmented intelligence, is inspired by recent developments in mathematics and physics. First applied tothe understanding of an important hyperparameter (temperature), then to a larger task (classification), theproposed method provides an intuition on the importance and role of factors correlated to the studied variable(e.g. hyperparameter, score, etc.). The processing chain thus set up has demonstrated its efficiency byproviding a highly explainable solution in line with decades of research in machine learning. These findings willallow the improvement of previously developed solutions
Paulin, Mathias. „Contributions à l'apprentissage automatique de réseau de contraintes et à la constitution automatique de comportements sensorimoteurs en robotique“. Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2008. http://tel.archives-ouvertes.fr/tel-00340438.
Der volle Inhalt der QuelleBredèche, Nicolas. „Ancrage de lexique et perceptions : changements de représentation et apprentissage dans le contexte d'un agent situé et mobile“. Paris 11, 2002. http://www.theses.fr/2002PA112225.
Der volle Inhalt der QuelleIn Artificial Intelligence, the symbol grounding problem is considered as an important issue regarding the meaning of symbols used by an artificial agent. Our work is concerned with the grounding of symbols for a situated mobile robot that navigates through a real world environment. In this setting, the main problem the robot encounters is to ground symbols given by a human teacher that refers to physical entities (e. G. A door, a human, etc. ). Grounding such a lexicon is a difficult task because of the intrinsic nature of the environment: it is dynamic, complex and noisy. Moreover, one specific symbol (e. G. "door") may refer to different physical objects in size, shape or colour while the robot may acquire only a small number of examples for each symbol. Also, it is not possible to rely on ad-hoc physical models of symbols due to the great number of symbols that may be grounded. Thus, the problem is to define how to build a grounded representation in such a context. In order to address this problem, we have reformulated the symbol grounding problem as a supervised learning problem. We present an approach that relies on the use of abstraction operators. Thanks to these operators, information on granularity and structural configuration is extracted from the perceptions in order to case the building of an anchor. For each symbol, the appropriate definition for these operators is found out thanks to successive changes of representation that provide an efficient and adapted anchor. In order to implement our approach, we have developed PLIC and WMplic which are successfully used for long term symbol grounding by a PIONEER2 DX mobile robot in the corridors of the Computer Sciences Lab of the University of Paris 6
Cogrel, Benjamin. „Sélection contextuelle de services continus pour la robotique ambiante“. Thesis, Paris Est, 2013. http://www.theses.fr/2013PEST1079/document.
Der volle Inhalt der QuelleAmbient robotics aims at introducing mobile robots in active environments where the latter provide new or alternative functionalities to those shipped by mobile robots. This thesis studies the competition between robot and external functionalities, which is set as a service selection problem. Service selection consists in choosing a service or a combination of services among a set of candidates able to fulfil a given request. To do this, it has to predict and evaluate candidate performances. These performances are based on non-functional requirements such as execution time, cost or noise. This application domain requires tight coordination between some of its functionalities. Tight coordination involves setting data streams between functionalities during their execution. In this proposal, functionalities producing data streams are modelled as continuous services. This new service category requires hierarchical service composition and adds some constraints to the service selection problem. This thesis shows that an important non-functional coupling appears between service instances at different levels, even when data streams are unidirectional. The proposed approach focuses on performance prediction of an high-level service instance given its organigram. This organigram gathers service instances involved in the high-level task processing. The scenario included in this study is the selection of a positioning service involved in a robot navigation high-level service. For a given organigram, performance prediction of an high-level service instance of this scenario has to: (i) be contextual by, for instance, considering moving path towards the required destination, (ii) support service instance replacement after a failure or in an opportunist manner. Consequently, this service selection is set as a sequential decision problem and is formalized as a finite-horizon Markov decision process. Its high contextual dimensionality with respect to robot moving frequency makes direct learning of Q-value functions or transition functions inadequate. The proposed approachre lies on local dynamic models and uses the planned moving path to estimate Q-values of organigrams available in the initial state. This estimation is done using a Monte-Carlo tree search method, Upper Confidence bounds applied to Trees
Hanna, Elias. „Improving Novelty Search Sample Efficiency with Models or Archive Bootstrapping“. Electronic Thesis or Diss., Sorbonne université, 2024. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2024SORUS038.pdf.
Der volle Inhalt der QuelleIn the context of policy search for robotic systems, being sample-efficient is a most. Evolutionary Algorithms have been used in the past ten years to achieve significant results in the robotics domain as their Darwinist approach to optimization allows them to bypass problems often encountered by gradient-based optimization methods like Reinforcement Learning. Nevertheless, such methods remain sample greedy and almost impossible to transfer directly onto a real robotic system. This Ph.D thesis takes interest in solving that sample-efficiency problem, especially for the Novelty Search algorithm, a novelty-driven Evolutionary Algorithm. Incorporating learned models in the optimization process has been a solution towards sample-efficiency for many years, but few works address this within the Novelty Search framework. Three research axis within that framework are explored in this manuscript. Firstly, the impact of pre-training the learned model with data gathered using random processes of varying time-correlation is evaluated. It is shown that the impact is negligible on a state-of-the-art model-based Evolutionary Algorithm, but that it is significant on a model-based Reinforcement Learning algorithm with returns with up to ten-fold differences between the best and worst random process used. Secondly, a preliminary study is made on a new approach aiming at biasing the initial population of the Novelty Search algorithm towards a more behavioraly diverse one using random dynamics models ensembles. It is shown that this approach successfully reduces the number of evaluations required by Novelty Search to cover the environment of a two-wheeled robot. It is also shown that this approach fails on a more complex locomotion environment of an hexapod robot, and the lack of diversity captured by the random models ensembles used is determined as the cause. Finally, a new model-based Evolutionary Algorithm, dubbed Model-Based Novelty Search, is proposed, with the aim of preserving the strong exploration capabilities of Novelty Search while reducing the numbers of evaluations needed to reach the same coverage of the Behavioral Space. Results on three robotic tasks show a reduction in sample usage compared to Novelty Search of 30% up to 75% depending on the considered task, a significant advance towards a more sample-efficient Novelty Search algorithm
Laflaquière, Alban. „Approche sensorimotrice de la perception de l'espace pour la robotique autonome“. Phd thesis, Université Pierre et Marie Curie - Paris VI, 2013. http://tel.archives-ouvertes.fr/tel-00865091.
Der volle Inhalt der QuellePerez, Manuela. „Chirurgie robotique : de l'apprentissage à l'application“. Thesis, Université de Lorraine, 2012. http://www.theses.fr/2012LORR0113/document.
Der volle Inhalt der QuelleThe huge expansion of minimally invasive robotic devices for surgery ask the question of the training of this new technology. Progress of robotic-assisted surgical techniques allows today mini- invasive surgery to be more accurate, providing benefits to surgeons and patients for complex surgical procedures. But, it resulted from an increasing need for training and development of new pedagogical strategies. Indeed, the surgeon has to master the telemanipulator and the procedure, which is different from a simple transposition of a laparoscopic skill. The first part of this work treats about historical development of minimal invasive surgery from laparoscopy to robotic surgery. We also develop the evolution of training program in surgery. Virtual simulators provide efficient tools for laparoscopy training. The second part of this work, study some possible solutions for robotic training. We assess the validity of a new robotic virtual simulator (dV-Trainer). We demonstrate the usefulness of this tool for the acquisition of the basic gesture for robotic surgery. Then, we evaluate the impact of a previous experience in micro-surgery on robotic training. We propose a prospective study comparing the surgical performance of micro-surgeons to that of general surgeons on a robotic simulator. We want to determine if this experience in micro-surgery could significantly improve the abilities and surgeons performance in the field of basic gesture in robotic surgery. The last part of the study also looks to the future. Currently, telesurgery need sophisticated dedicated technical resources. We want to develop procedures for clinical routine used. Therefore, we evaluate the impact of the delay on the surgical procedure. Also, reducing data volume allow decreasing latency. An appropriate solution to reduce the amount of data could be found by introducing lossy compression for the transmission using the well-known MPEG-2 and H-264 standards
Ramoly, Nathan. „Contextual integration of heterogeneous data in an open and opportunistic smart environment : application to humanoid robots“. Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLL003/document.
Der volle Inhalt der QuellePersonal robots associated with ambient intelligence are an upcoming solution for domestic care. In fact, helped with devices dispatched in the environment, robots could provide a better care to users. However, such robots are encountering challenges of perception, cognition and action.In fact, such an association brings issues of variety, data quality and conflicts, leading to the heterogeneity and uncertainty of data. These are challenges for both perception, i.e. context acquisition, and cognition, i.e. reasoning and decision making. With the knowledge of the context, the robot can intervene through actions. However, it may encounter task failures due to a lack of knowledge or context changes. This causes the robot to cancel or delay its agenda. While the literature addresses those topics, it fails to provide complete solutions. In this thesis, we proposed contributions, exploring both reasoning and learning approaches, to cover the whole spectrum of problems. First, we designed novel context acquisition tool that supports and models uncertainty of data. Secondly, we proposed a cognition technique that detects anomalous situation over uncertain data and takes a decision in accordance. Then, we proposed a dynamic planner that takes into consideration the last context changes. Finally, we designed an experience-based reinforcement learning approach to proactively avoid failures.All our contributions were implemented and validated through simulations and/or with a small robot in a smart home platform
Dermy, Oriane. „Prédiction du mouvement humain pour la robotique collaborative : du geste accompagné au mouvement corps entier“. Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0227/document.
Der volle Inhalt der QuelleThis thesis lies at the intersection between machine learning and humanoid robotics, under the theme of human-robot interaction and within the cobotics (collaborative robotics) field. It focuses on prediction for non-verbal human-robot interactions, with an emphasis on gestural interaction. The prediction of the intention, understanding, and reproduction of gestures are therefore central topics of this thesis. First, the robots learn gestures by demonstration: a user grabs its arm and makes it perform the gestures to be learned several times. The robot must then be able to reproduce these different movements while generalizing them to adapt them to the situation. To do so, using its proprioceptive sensors, it interprets the perceived signals to understand the user's movement in order to generate similar ones later on. Second, the robot learns to recognize the intention of the human partner based on the gestures that the human initiates. The robot can then perform gestures adapted to the situation and corresponding to the user’s expectations. This requires the robot to understand the user’s gestures. To this end, different perceptual modalities have been explored. Using proprioceptive sensors, the robot feels the user’s gestures through its own body: it is then a question of physical human-robot interaction. Using visual sensors, the robot interprets the movement of the user’s head. Finally, using external sensors, the robot recognizes and predicts the user’s whole body movement. In that case, the user wears sensors (in our case, a wearable motion tracking suit by XSens) that transmit his posture to the robot. In addition, the coupling of these modalities was studied. From a methodological point of view, the learning and the recognition of time series (gestures) have been central to this thesis. In that aspect, two approaches have been developed. The first is based on the statistical modeling of movement primitives (corresponding to gestures) : ProMPs. The second adds Deep Learning to the first one, by using auto-encoders in order to model whole-body gestures containing a lot of information while allowing a prediction in soft real time. Various issues were taken into account during this thesis regarding the creation and development of our methods. These issues revolve around: the prediction of trajectory durations, the reduction of the cognitive and motor load imposed on the user, the need for speed (soft real-time) and accuracy in predictions
Munzer, Thibaut. „Représentations relationnelles et apprentissage interactif pour l'apprentissage efficace du comportement coopératif“. Thesis, Bordeaux, 2017. http://www.theses.fr/2017BORD0574/document.
Der volle Inhalt der QuelleThis thesis presents new approaches toward efficient and intuitive high-level plan learning for cooperative robots. More specifically this work study Learning from Demonstration algorithm for relational domains. Using relational representation to model the world, simplify representing concurrentand cooperative behavior.We have first developed and studied the first algorithm for Inverse ReinforcementLearning in relational domains. We have then presented how one can use the RAP formalism to represent Cooperative Tasks involving a robot and a human operator. RAP is an extension of the Relational MDP framework that allows modeling concurrent activities. Using RAP allow us to represent both the human and the robot in the same process but also to model concurrent robot activities. Under this formalism, we have demonstrated that it is possible to learn behavior, as policy and as reward, of a cooperative team. Prior knowledge about the task can also be used to only learn preferences of the operator.We have shown that, using relational representation, it is possible to learn cooperative behaviors from a small number of demonstration. That these behaviors are robust to noise, can generalize to new states and can transfer to different domain (for example adding objects). We have also introduced an interactive training architecture that allows the system to make fewer mistakes while requiring less effort from the human operator. By estimating its confidence the robot is able to ask for instructions when the correct activity to dois unsure. Lastly, we have implemented these approaches on a real robot and showed their potential impact on an ecological scenario