Autonomous mobile robots are being developed with the aim of accomplishing complex tasks in different environments, including human habitats as well as less friendly places, such as distant planets and underwater regions. A major challenge faced by such robots is to make sure that their actions are executed correctly and reliably, despite the dynamics and the uncertainty inherent in their working space. This thesis is concerned with the ability of a mobile robot to reliably monitor the execution of its plans and detect failures. Existing approaches for monitoring the execution of plans rely mainly on checking the explicit effects of plan actions, i.e., effects encoded in the action model. This supposedly means that the effects to monitor are directly observable, but that is not always the case in a real-world environment. In this thesis, we propose to use semantic domain-knowledge to derive and monitor implicit expectations about the effects of actions. For instance, a robot entering a room asserted to be an office should expect to see at least a desk, a chair, and possibly a PC. These expectations are derived from knowledge about the type of the room the robot is entering. If the robot enters a kitchen instead, then it should expect to see an oven, a sink, etc. The major contributions of this thesis are as follows. • We define the notion of Semantic Knowledge-based Execution Monitoring SKEMon, and we propose a general algorithm for it based on the use of description logics for representing knowledge. • We develop a probabilistic approach of semantic Knowledge-based execution monitoring to take into account uncertainty in both acting and sensing. Specifically, we allow for sensing to be unreliable and for action models to have more than one possible outcome. We also take into consideration uncertainty about the state of the world. This development is essential to the applicability of our technique, since uncertainty is a pervasive feature in robotics. • We present a general schema to deal with situations where perceptual information relevant to SKEMon is missing. The schema includes steps for modeling and generating a course of action to actively collect such information. We describe approaches based on planning and greedy action selection to generate the information-gathering solutions. The thesis also shows how such a schema can be applied to respond to failures occurring before or while an action is executed. The failures we address are ambiguous situations that arise when the robot attempts to anchor symbolic descriptions (relevant to a plan action) in perceptual information. The work reported in this thesis has been tested and verified using a mobile robot navigating in an indoor environment. In addition, simulation experiments were conducted to evaluate the performance of SKEMon using known metrics. The results show that using semantic knowledge can lead to high performance in monitoring the execution of robot plans.

Negation is a part of language that humans engage in pretty much from the onset of speech. Negation appears at first glance to be harder to grasp than object or action labels, yet this thesis explores how this family of ‘concepts’ could be acquired in a meaningful way by a humanoid robot based solely on the unconstrained dialogue with a human conversation partner. The earliest forms of negation appear to be linked to the affective or motivational state of the speaker. Therefore we developed a behavioural architecture which contains a motivational system. This motivational system feeds its state simultaneously to other subsystems for the purpose of symbol-grounding but also leads to the expression of the robot’s motivational state via a facial display of emotions and motivationally congruent body behaviours. In order to achieve the grounding of negative words we will examine two different mechanisms which provide an alternative to the established grounding via ostension with or without joint attention. Two large experiments were conducted to test these two mechanisms. One of these mechanisms is so called negative intent interpretation, the other one is a combination of physical and linguistic prohibition. Both mechanisms have been described in the literature on early child language development but have never been used in human-robot-interaction for the purpose of symbol grounding. As we will show, both mechanisms may operate simultaneously and we can exclude none of them as potential ontogenetic origin of negation.

Si la question de l'origine du langage reste d'un abord compliqué, celle de l'origine des formes du langage semble plus susceptible de se confronter à la démarche expérimentale. Malgré leur infinie variété, d'évidentes régularités y sont présentes~: les universaux du langage. Nous les étudions par des raisonnements plus généraux sur l'émergence du langage, notamment sur la recherche de précurseurs onto- et phylogénétiques. Nous abordons trois thèmes principaux~: la situation de communication parlée, les architectures cognitives des agents et l'émergence des universaux du langage dans des sociétés d'agents. Notre première contribution est un modèle conceptuel des agents communicants en interaction, issu de notre analyse bibliographique. Nous en proposons ensuite une formalisation mathématique Bayésienne~: le modèle d'un agent est une distribution de probabilités, et la production et la perception sont des inférences bayésiennes. Cela permet la comparaison formelle des différents courants théoriques en perception et en production de la parole. Enfin, nos simulations informatiques de société d'agents identifient les conditions qui favorisent l'apparition des universaux du langage. / If the origin of language is difficult to properly study, the origin of its forms appears to be accessible to the experimental method. Languages, despite their large variety, display obvious regularities, the linguistic universals. We study them through more general reasoning about language emergence, in particular in the search of its precursors, both in ontogeny and phylogeny. We study three main themes: the communication situation, the agent's cognitive architectures and the emergence of linguistic universals in agent societies. Our first contribution is a conceptual model of communicating agents in interaction, emanating from our bibliographic survey. We then cast it into the Bayesian mathematical formalism: an agent model is a probability distribution, and production and perception are defined by Bayesian inference. This allows a formal comparison of speech perception and production theoretical trends. Finally, computer simulations of agent societies help identify the conditions that favor the appearance of linguistic universals.

Building general purpose autonomous robots that suit a wide range of user-specified applications, requires a leap from today's task-specific machines to more flexible and general ones. To achieve this goal, one should move from traditional preprogrammed robots to learning robots that easily can acquire new skills. Learning from Demonstration (LfD) and Imitation Learning (IL), in which the robot learns by observing a human or robot tutor, are among the most popular learning techniques. Showing the robot how to perform a task is often more natural and intuitive than figuring out how to modify a complex control program. However, teaching robots new skills such that they can reproduce the acquired skills under any circumstances, on the right time and in an appropriate way, require good understanding of all challenges in the field. Studies of imitation learning in humans and animals show that several cognitive abilities are engaged to learn new skills correctly. The most remarkable ones are the ability to direct attention to important aspects of demonstrations, and adapting observed actions to the agents own body. Moreover, a clear understanding of the demonstrator's intentions and an ability to generalize to new situations are essential. Once learning is accomplished, various stimuli may trigger the cognitive system to execute new skills that have become part of the robot's repertoire. The goal of this thesis is to develop methods for learning from demonstration that mainly focus on understanding the tutor's intentions, and recognizing which elements of a demonstration need the robot's attention. An architecture containing required cognitive functions for learning and reproduction of high-level aspects of demonstrations is proposed. Several learning methods for directing the robot's attention and identifying relevant information are introduced. The architecture integrates motor actions with concepts, objects and environmental states to ensure correct reproduction of skills. Another major contribution of this thesis is methods to resolve ambiguities in demonstrations where the tutor's intentions are not clearly expressed and several demonstrations are required to infer intentions correctly. The provided solution is inspired by human memory models and priming mechanisms that give the robot clues that increase the probability of inferring intentions correctly. In addition to robot learning, the developed techniques are applied to a shared control system based on visual servoing guided behaviors and priming mechanisms. The architecture and learning methods are applied and evaluated in several real world scenarios that require clear understanding of intentions in the demonstrations. Finally, the developed learning methods are compared, and conditions where each of them has better applicability are discussed. / Att bygga autonoma robotar som passar ett stort antal olika användardefinierade applikationer kräver ett språng från dagens specialiserade maskiner till mer flexibla lösningar. För att nå detta mål, bör man övergå från traditionella förprogrammerade robotar till robotar som själva kan lära sig nya färdigheter. Learning from Demonstration (LfD) och Imitation Learning (IL), där roboten lär sig genom att observera en människa eller en annan robot, är bland de mest populära inlärningsteknikerna. Att visa roboten hur den ska utföra en uppgift är ofta mer naturligt och intuitivt än att modifiera ett komplicerat styrprogram. Men att lära robotar nya färdigheter så att de kan reproducera dem under nya yttre förhållanden, på rätt tid och på ett lämpligt sätt, kräver god förståelse för alla utmaningar inom området. Studier av LfD och IL hos människor och djur visar att flera kognitiva förmågor är inblandade för att lära sig nya färdigheter på rätt sätt. De mest anmärkningsvärda är förmågan att rikta uppmärksamheten på de relevanta aspekterna i en demonstration, och förmågan att anpassa observerade rörelser till robotens egen kropp. Dessutom är det viktigt att ha en klar förståelse av lärarens avsikter, och att ha förmågan att kunna generalisera dem till nya situationer. När en inlärningsfas är slutförd kan stimuli trigga det kognitiva systemet att utföra de nya färdigheter som blivit en del av robotens repertoar. Målet med denna avhandling är att utveckla metoder för LfD som huvudsakligen fokuserar på att förstå lärarens intentioner, och vilka delar av en demonstration som ska ha robotens uppmärksamhet. Den föreslagna arkitekturen innehåller de kognitiva funktioner som behövs för lärande och återgivning av högnivåaspekter av demonstrationer. Flera inlärningsmetoder för att rikta robotens uppmärksamhet och identifiera relevant information föreslås. Arkitekturen integrerar motorkommandon med begrepp, föremål och omgivningens tillstånd för att säkerställa korrekt återgivning av beteenden. Ett annat huvudresultat i denna avhandling rör metoder för att lösa tvetydigheter i demonstrationer, där lärarens intentioner inte är klart uttryckta och flera demonstrationer är nödvändiga för att kunna förutsäga intentioner på ett korrekt sätt. De utvecklade lösningarna är inspirerade av modeller av människors minne, och en primingmekanism används för att ge roboten ledtrådar som kan öka sannolikheten för att intentioner förutsägs på ett korrekt sätt. De utvecklade teknikerna har, i tillägg till robotinlärning, använts i ett halvautomatiskt system (shared control) baserat på visuellt guidade beteenden och primingmekanismer. Arkitekturen och inlärningsteknikerna tillämpas och utvärderas i flera verkliga scenarion som kräver en tydlig förståelse av mänskliga intentioner i demonstrationerna. Slutligen jämförs de utvecklade inlärningsmetoderna, och deras applicerbarhet under olika förhållanden diskuteras. / INTRO

Um sistema artificial pode usar raciocínio espacial qualitativo para inferir informações sobre seu ambiente tridimensional a partir de imagens bidimensionais. Inferências realizadas com base em raciocínio espacial qualitativo devem ser capazes de lidar com incertezas. Neste trabalho investigamos a utilização de técnicas probabilísticas para tornar o raciocínio espacial qualitativo mais robusto a incertezas e aplicável a agentes móveis em ambientes reais. Investigamos uma formalização de raciocínio espacial com lógica de descrição probabilística em um subdomínio de tráfego. Desenvolvemos também um método que combina raciocínio espacial qualitativo com um filtro Bayesiano para desenvolver dois sistemas que foram aplicados na auto localização de um robô móvel. Executamos dois experimentos de auto localização; um utilizando a teoria de relações qualitativas percebíveis sobre sombra com filtro Bayesiano; e outro utilizando o cálculo de oclusão de regiões e o cálculo de direção com filtro Bayesiano. Ambos os sistemas obtiveram resultados positivos onde somente o raciocínio espacial qualitativo não foi capaz de inferir a localização do robô. Os experimentos com dados reais mostraram robustez aos ruídos e à informação parcial. / An artificial system can use qualitative spatial reasoning to obtain information about its tridimensional environment, from bi-dimensional images. Inferences produced by qualitative spatial reasoning must be able to deal with uncertainty. This work investigates the use of probabilistic techniques to make qualitative spatial reasoning more robust against uncertainty, and better applicable to mobile agents in real environments. The work investigates a formalization of spatial reasoning using probabilistic description logics in a traffic domain. Additionally, a method is presented that combines qualitative spatial reasoning with a Bayesian filter, to develop two systems that are applied to self-localization of mobile robots. Two experiments are described; one using the theory of perceptual qualitative relations about shadows; the other using occlusion calculus and direction calculus. Both systems are combined with a Bayesian filter producing positive results in situations where qualitative spatial reasoning alone cannot infer robot location. Experiments with real data show robustness to noise and partial information.

Um sistema artificial pode usar raciocínio espacial qualitativo para inferir informações sobre seu ambiente tridimensional a partir de imagens bidimensionais. Inferências realizadas com base em raciocínio espacial qualitativo devem ser capazes de lidar com incertezas. Neste trabalho investigamos a utilização de técnicas probabilísticas para tornar o raciocínio espacial qualitativo mais robusto a incertezas e aplicável a agentes móveis em ambientes reais. Investigamos uma formalização de raciocínio espacial com lógica de descrição probabilística em um subdomínio de tráfego. Desenvolvemos também um método que combina raciocínio espacial qualitativo com um filtro Bayesiano para desenvolver dois sistemas que foram aplicados na auto localização de um robô móvel. Executamos dois experimentos de auto localização; um utilizando a teoria de relações qualitativas percebíveis sobre sombra com filtro Bayesiano; e outro utilizando o cálculo de oclusão de regiões e o cálculo de direção com filtro Bayesiano. Ambos os sistemas obtiveram resultados positivos onde somente o raciocínio espacial qualitativo não foi capaz de inferir a localização do robô. Os experimentos com dados reais mostraram robustez aos ruídos e à informação parcial. / An artificial system can use qualitative spatial reasoning to obtain information about its tridimensional environment, from bi-dimensional images. Inferences produced by qualitative spatial reasoning must be able to deal with uncertainty. This work investigates the use of probabilistic techniques to make qualitative spatial reasoning more robust against uncertainty, and better applicable to mobile agents in real environments. The work investigates a formalization of spatial reasoning using probabilistic description logics in a traffic domain. Additionally, a method is presented that combines qualitative spatial reasoning with a Bayesian filter, to develop two systems that are applied to self-localization of mobile robots. Two experiments are described; one using the theory of perceptual qualitative relations about shadows; the other using occlusion calculus and direction calculus. Both systems are combined with a Bayesian filter producing positive results in situations where qualitative spatial reasoning alone cannot infer robot location. Experiments with real data show robustness to noise and partial information.

La conception de systèmes de perception autonomes, tels que des robots capables d’accomplir un ensemble de tâches de manière sûre et sans assistance humaine, est l’un des grands défis de notre siècle. Pour ce faire, la robotique développementale propose de concevoir des robots qui, comme des enfants, auraient la faculté d’apprendre directement par interaction avec leur environnement. Nous avons dans cette thèse exploré de telles possibilités en se limitant à l’apprentissage de la localisation des objets d’intérêt (ou objets saillants) dans l’environnement du robot.Pour ce faire, nous présentons dans ces travaux un mécanisme capable d’apprendre la saillance visuelle directement sur un robot, puis d’utiliser le modèle appris de la sorte pour localiser des objets saillants dans son environnement. Cette méthode a l’avantage de permettre la création de modèles spécialisés pour l’environnement du robot et les tâches qu’il doit accomplir, tout en restant flexible à d’éventuelles nouveautés ou modifications de l’environnement.De plus, afin de permettre un apprentissage efficace et de qualité, nous avons développé des stratégies d’explorations basées sur les motivations intrinsèques, très utilisées en robotique développementale. Nous avons notamment adapté l’algorithme IAC à l’apprentissage de la saillance visuelle, et en avons conçu une extension, RL-IAC, pour permettre une exploration efficace sur un robot mobile. Afin de vérifier et d’analyser les performances de nos algorithmes, nous avons réalisé des évaluations sur plusieurs plateformes robotiques dont une plateforme fovéale et un robot mobile, ainsi que sur des bases de données publiques. / Conceiving autonomous perceptual systems, such as robots able to accomplish a set of tasks in a safe way, without any human assistance, is one of the biggest challenge of the century. To this end, the developmental robotics suggests to conceive robots able to learn by interacting directly with their environment, just like children would. This thesis is exploring such possibility while restricting the problem to the one of localizing objects of interest (or salient objects) within the robot’s environment.For that, we present in this work a mechanism able to learn visual saliency directly on a robot, then to use the learned model so as to localize salient objects within their environment. The advantage of this method is the creation of models dedicated to the robot’s environment and tasks it should be asked to accomplish, while remaining flexible to any change or novelty in the environment.Furthermore, we have developed exploration strategies based on intrinsic motivations, widely used in developmental robotics, to enable efficient learning of good quality. In particular, we adapted the IAC algorithm to visual saliency leanring, and proposed an extension, RL-IAC to allow an efficient exploration on mobile robots.In order to verify and analyze the performance of our algorithms, we have carried out various experiments on several robotics platforms, including a foveated system and a mobile robot, as well as publicly available datasets.

Avec le développement de la robotique cognitive, le besoin d’outils avancés pour représenter, manipuler, raisonner sur les connaissances acquises par un robot a clairement été mis en avant. Mais stocker et manipuler des connaissances requiert tout d’abord d’éclaircir ce que l’on nomme connaissance pour un robot, et comment celle-ci peut-elle être représentée de manière intelligible pour une machine. Ce travail s’efforce dans un premier temps d’identifier de manière systématique les besoins en terme de représentation de connaissance des applications robotiques modernes, dans le contexte spécifique de la robotique de service et des interactions homme-robot. Nous proposons une typologie originale des caractéristiques souhaitables des systèmes de représentation des connaissances, appuyée sur un état de l’art détaillé des outils existants dans notre communauté. Dans un second temps, nous présentons en profondeur ORO, une instanciation particulière d’un système de représentation et manipulation des connaissances, conçu et implémenté durant la préparation de cette thèse. Nous détaillons le fonctionnement interne du système, ainsi que son intégration dans plusieurs architectures robotiques complètes. Un éclairage particulier est donné sur la modélisation de la prise de perspective dans le contexte de l’interaction, et de son interprétation en terme de théorie de l’esprit. La troisième partie de l’étude porte sur une application importante des systèmes de représentation des connaissances dans ce contexte de l’interaction homme-robot : le traitement du dialogue situé. Notre approche et les algorithmes qui amènent à l’ancrage interactif de la communication verbale non contrainte sont présentés, suivis de plusieurs expériences menées au Laboratoire d’Analyse et d’Architecture des Systèmes au CNRS à Toulouse, et au groupe Intelligent Autonomous System de l’université technique de Munich. Nous concluons cette thèse sur un certain nombre de considérations sur la viabilité et l’importance d’une gestion explicite des connaissances des agents, ainsi que par une réflexion sur les éléments encore manquant pour réaliser le programme d’une robotique “de niveau humain” / With the rise of the so-called cognitive robotics, the need of advanced tools to store, manipulate, reason about the knowledge acquired by the robot has been made clear. But storing and manipulating knowledge requires first to understand what the knowledge itself means to the robot and how to represent it in a machine-processable way. This work strives first at providing a systematic study of the knowledge requirements of modern robotic applications in the context of service robotics and human-robot interaction. What are the expressiveness requirement for a robot? what are its needs in term of reasoning techniques? what are the requirement on the robot's knowledge processing structure induced by other cognitive functions like perception or decision making? We propose a novel typology of desirable features for knowledge representation systems supported by an extensive review of existing tools in our community. In a second part, the thesis presents in depth a particular instantiation of a knowledge representation and manipulation system called ORO, that has been designed and implemented during the preparation of the thesis. We elaborate on the inner working of this system, as well as its integration into several complete robot control stacks. A particular focus is given to the modelling of agent-dependent symbolic perspectives and their relations to theories of mind. The third part of the study is focused on the presentation of one important application of knowledge representation systems in the human-robot interaction context: situated dialogue. Our approach and associated algorithms leading to the interactive grounding of unconstrained verbal communication are presented, followed by several experiments that have taken place both at the Laboratoire d'Analyse et d'Architecture des Systèmes at CNRS, Toulouse and at the Intelligent Autonomous System group at Munich Technical University. The thesis concludes on considerations regarding the viability and importance of an explicit management of the agent's knowledge, along with a reflection on the missing bricks in our research community on the way towards "human level robots"

This dissertation investigates high-level decision making for agents that are both goal and utility driven. We develop a partially observable Markov decision process (POMDP) planner which is an extension of an agent programming language called DTGolog, itself an extension of the Golog language. Golog is based on a logic for reasoning about action—the situation calculus. A POMDP planner on its own cannot cope well with dynamically changing environments and complicated goals. This is exactly a strength of the belief-desire-intention (BDI) model: BDI theory has been developed to design agents that can select goals intelligently, dynamically abandon and adopt new goals, and yet commit to intentions for achieving goals. The contribution of this research is twofold: (1) developing a relational POMDP planner for cognitive robotics, (2) specifying a preliminary BDI architecture that can deal with stochasticity in action and perception, by employing the planner. / Computing / M. Sc. (Computer Science)

This dissertation investigates high-level decision making for agents that are both goal and utility driven. We develop a partially observable Markov decision process (POMDP) planner which is an extension of an agent programming language called DTGolog, itself an extension of the Golog language. Golog is based on a logic for reasoning about action—the situation calculus. A POMDP planner on its own cannot cope well with dynamically changing environments and complicated goals. This is exactly a strength of the belief-desire-intention (BDI) model: BDI theory has been developed to design agents that can select goals intelligently, dynamically abandon and adopt new goals, and yet commit to intentions for achieving goals. The contribution of this research is twofold: (1) developing a relational POMDP planner for cognitive robotics, (2) specifying a preliminary BDI architecture that can deal with stochasticity in action and perception, by employing the planner. / Computing / M. Sc. (Computer Science)