For robots to operate in real life settings, they must be able to physically interact with the environment, and for instance be able to react to force-guidance interactions. However, only a few research projects have addressed such capabilities, developing prototypes that have to be pushed from their handle bars. AZIMUT-3 is a novel omnidirectional non-holonomic mobile robot developed at IntRoLab (Intelligent, Interactive and Interdisciplinary Robot Lab, Université de Sherbrooke) with force-controlled active steering. This results in a horizontal suspension effect for which the mechanical impedance of the steering actuators can be controlled. This makes the platform ideal for developing physical guidance algorithms. One such algorithm is secured shared-control, making the platform go in the direction of the user pushing the robot while still making it move safely by avoiding obstacles. Such capability is somewhat novel in the field, and the objective is to provide safe navigation with maximum control to the user. This Master's thesis has two important contributions: an algorithm to estimate the applied efforts on AZIMUT-3 from torque measurements on its wheels; an algorithm to use these efforts with obstacle detection using laser range finder data to implement a safe, shared-control approach. Experimental results using the real platform demonstrate feasibility and safe control of the system, with performances similar to using a six degrees of freedom force sensor but at lower cost and with a broader area for shared control. Our implementation also resulted in coupling the simulation environment Webots with the ROS (Robot Operating System) library from Willow Garage, to help develop our approach in simulation before using AZIMUT-3. Overall, our work is the first in demonstrating how it is possible to naturally interact by physically moving or positioning a mobile platform in real life settings, a capability which could be useful for instance in the design of powered shopping carts or active walkers.

International Telemetering Conference Proceedings / October 21, 2002 / Town & Country Hotel and Conference Center, San Diego, California / This paper describes the construction of an autonomous soccer playing robot as part of a senior design project at Brigham Young University. Each participating team designed and built a robot to compete in an annual tournament. To accomplish this, each team had access to images received from a camera placed above a soccer field. The creation of image processing and artificial intelligence software were required to allow the robot to perform against other robots in a one-on-one competition. Each participating team was given resources to accomplish this project. This paper contains a summary of the experiences gained by team members and also a description of the key components created for the robot named Prometheus to compete and win the annual tournament.

Robust long-term positioning for autonomous mobile robots is essential for many applications. In many environments this task is challenging, as errors accumulate in the robot’s position estimate over time. The robot must also build a map so that these errors can be corrected when mapped regions are re-visited; this is known as Simultaneous Localisation and Mapping, or SLAM. Successful SLAM schemes have been demonstrated which accurately map tracks of tens of kilometres, however these schemes rely on expensive sensors such as laser scanners and inertial measurement units. A more attractive, low-cost sensor is a digital camera, which captures images that can be used to recognise where the robot is, and to incrementally position the robot as it moves. SLAM using a single camera is challenging however, and many contemporary schemes suffer complete failure in dynamic or featureless environments, or during erratic camera motion. An additional problem, known as scale drift, is that cameras do not directly measure the scale of the environment, and errors in relative scale accumulate over time, introducing errors into the robot’s speed and position estimates. Key to a successful visual SLAM system is the ability to continue operation despite these difficulties, and to recover from positioning failure when it occurs. This thesis describes the development of such a scheme, which is known as BoWSLAM. BoWSLAM enables a robot to reliably navigate and map previously unknown environments, in real-time, using only a single camera. In order to position a camera in visually challenging environments, BoWSLAM combines contemporary visual SLAM techniques with four new components. Firstly, a new Bag-of-Words (BoW) scheme is developed, which allows a robot to recognise places it has visited previously, without any prior knowledge of its environment. This BoW scheme is also used to select the best set of frames to reconstruct positions from, and to find efficient wide-baseline correspondences between many pairs of frames. Secondly, BaySAC, a new outlier- robust relative pose estimation scheme based on the popular RANSAC framework, is developed. BaySAC allows the efficient computation of multiple position hypotheses for each frame. Thirdly, a graph-based representation of these position hypotheses is proposed, which enables the selection of only reliable position estimates in the presence of gross outliers. Fourthly, as the robot explores, objects in the world are recognised and measured. These measurements enable scale drift to be corrected. BoWSLAM is demonstrated mapping a 25 minute 2.5km trajectory through a challenging and dynamic outdoor environment in real-time, and without any other sensor input; considerably further than previous single camera SLAM schemes.

Although humanoid robots are being designed to assist people in various tasks, there remain gaps in our understanding of the perceptions that humanoid faces evoke in the user. Understanding user perceptions would help design robots that are better suited for the target user group. Younger and older adults’ preferences for robot appearance were assessed out of three levels of human-likeness. In general, people perceived a mixed human-robot appearance less favorably compared to highly human and highly robotic appearances. Additionally the nature of task also influenced people’s overall perceptions of robots. Robots were most positively evaluated for assistance with chores and less positively for personal care and decision-making. Moreover, task and robot humanness had an interactive effect on people’s likability, trust, and perceived usefulness toward robots. Age-related differences in preferences of robot humanness were also observed. Older adults showed a higher inclination toward human-looking appearance of robots whereas younger adults’ preferences were more distributed across the levels of humanness. An appearance with mixed human-robot features was more likely to be rejected by older adults than by younger adults, and the difference was most striking for a decision-making task. Besides the humanness of the robot face, perceptions of robot appearances were also influenced by factors such as robot gender, specific facial features/aesthetics, expressiveness, perceived personality, and perceived capability. Future studies should measure the relative weight of these different factors in the formation of perceptions, both at a global level and at a task-specific level.

This thesis is part of the Aurora project, an ongoing long-term project investigating the potential use of robots to help children with autism overcome some of their impairments in social interaction, communication and imagination. Autism is a spectrum disorder and children with autism have different abilities and needs. Related research has shown that robots can play the role of a mediator for social interaction in the context of autism. Robots can enable simple interactions, by initially providing a relatively predictable environment for play. Progressively, the complexity of the interaction can be increased. The purpose of this thesis is to facilitate play between children with autism and an autonomous robot. Children with autism have a potential for play but often encounter obstacles to actualize this potential. Through play, children can develop multidisciplinary skills, involving social interaction, communication and imagination. Besides, play is a medium for self-expression. The purpose here is to enable children with autism to experience a large range of play situations, ranging from dyadic play with progressively better balanced interaction styles, to situations of triadic play with both the robot and the experimenter. These triadic play situations could also involve symbolic or pretend play. This PhD work produced the following results: • A new methodological approach of how to design, conduct and analyse robotassisted play was developed and evaluated. This approach draws inspiration from non-directive play therapy where the child is the main leader for play and the experimenter participates in the play sessions. I introduced a regulation process which enables the experimenter to intervene under precise conditions in order to: i) prevent the child from entering or staying in repetitive behaviours, ii) provide bootstrapping that helps the child reach a situation of play she is about to enter and iii) ask the child questions dealing with affect or reasoning about the robot. This method has been tested in a long-term study with six children with autism. Video recordings of the play sessions were analysed in detail according to three dimensions, namely Play, Reasoning and Affect. Results have shown the ability of this approach to meet each child’s specific needs and abilities. Future work may develop this work towards a novel approach in autism therapy. • A novel and generic computational method for the automatic recognition of human-robot interaction styles (specifically gentleness and frequency of touch interaction) in real time was developed and tested experimentally. This method, the Cascaded Information Bottleneck Method, is based on an information theoretic approach. It relies on the principle that the relevant information can be progressively extracted from a time series with a cascade of successive bottlenecks sharing the same cardinality of bottleneck states but trained successively. This method has been tested with data that had been generated with a physical robot a) during human-robot interactions in laboratory conditions and b) during child-robot interactions in school. The method shows a sound recognition of both short-term and mid-term time scale events. The recognition process only involves a very short delay. The Cascaded Information Bottleneck is a generic method that can potentially be applied to various applications of socially interactive robots. • A proof-of-concept system of an adaptive robot was demonstrated that is responsive to different styles of interaction in human-robot interaction. Its impact was evaluated in a short-term study with seven children with autism. The recognition process relies on the Cascaded Information Bottleneck Method. The robot rewards well-balanced interaction styles. The study shows the potential of the adaptive robot i) to encourage children to engage more in the interaction and ii) to positively influence the children’s play styles towards better balanced interaction styles. It is hoped that this work is a step forward towards socially adaptive robots as well as robot-assisted play for children with autism.

L’apparition de robots de service de plus en plus complexes ouvre de nouvelles perspectives quant aux tâches de manipulation d’objets. Malgré les progrès récents des techniques de planification de mouvement, peu d'entre elles s'intéressent directement à des systèmes multi-bras comme les torses humanoïdes. Notre contribution à travers cette thèse porte sur trois aspects. Nous proposons une technique de planification de mouvement performante basée sur la coordination des mouvements du système multi-bras. Elle exploite au mieux la structure du système en la divisant en parties élémentaires dont les mouvements sont planifiés indépendamment du reste du système. La fusion des différents réseaux élémentaires générés est ensuite réalisée dans le but d’obtenir un graphe prenant en compte le robot tout entier. Une seconde contribution porte sur l'extension des méthodes de planification pour des robots présentant des chaînes cinématiques fermées. Ces boucles cinématiques apparaissent dans le système lorsque, par exemple, le torse humanoïde saisit un objet avec plusieurs bras. Cette méthode traite explicitement les configurations singulières des manipulateurs, offrant ainsi une meilleure maniabilité de l’objet. Finalement, nous proposons deux approches pour la planification de tâches de manipulation d'objets par un torse humanoïde. La première concerne la résolution d’une tâche de prise et pose d'objets par un torse humanoïde à deux bras dans le cas où les contraintes imposées par la tâche nécessitent le passage par une double prise afin de transférer l'objet d'une main à l'autre. La seconde porte sur la résolution du même type de tâche par un manipulateur mobile. La thèse, effectuée dans le cadre du projet européen Phriends, présente les résultats d'expérimentations réalisées sur le robot Justin, démonstrateur du projet. / The emergence of new more and more complex service robots opens new research fields on objet manipulation. Despite the recent progresses in motion planning techniques, few of them deal directly with multi-arm systems like humanoid torsos. Our contribution through this thesis focuses on three aspects. We present an efficient motion planning technique based on the multi-arm system motion coordination. It takes advantage of the system's structure by dividing it into elementary parts of which movements are planned independently of the rest of the system. Generated elementary networks are then fused to obtain a roadmap that takes into account the whole robot. The second contribution consists of the extension of motion planning methods for a robot under loop closure constraints. These kinematic loops appear in the system when, for example, the humanoid torso grasps an objet with two arms. This method treats explicitly the singular configurations of the manipulators, providing better handling of the object. Finally, we present two approaches for planning object manipulation tasks by humanoid torsos. The first concerns solving pick and place task by humanoid torso where the imposed task constraints require a passage through a double grasp to transfer the object from one hand to the other. The second approach concerns the resolution of the same type of task by a mobile manipulator. The presented methods have been integrated on a real platform, Justin, and validated with experiments in the frame of E.U. FP-6 PHRIENDS project.

Les architectures logicielles de contrôles sont le centre névralgique des robots. Malheureusement les robots et leurs architectures souffrent de nombreuses imperfections qui perturbent et/ou compromettent la réalisation des missions qui leurs sont affectés. Nous proposons donc une méthodologie de conception d'architecture de contrôle adaptative pour la mise en œuvre de la tolérance aux fautes.La première partie de ce manuscrit propose un état de l'art de la sureté de fonctionnement, d'abord générique avant d'être spécifié au contexte des architectures de contrôle. La seconde partie nous permet de détailler la méthodologie proposée permettant d'identifier les fautes potentielles d'un robot et d'y répondre à l'aide des moyens de tolérance aux fautes. La troisième partie présente le contexte expérimental et applicatif dans lequel la méthodologie proposée sera mise en œuvre et qui constitue la quatrième partie de ce manuscrit. Une expérimentation spécifique mettant en lumière les aspects de la méthodologie est détaillée dans la dernière partie. / The software control architectures are the decisional center of robots. Unfortunately, the robots and their architectures suffer from numerous flaws that disrupt and / or compromise the achievement of missions they are assigned. We therefore propose a methodology for designing adaptive control architecture for the implementation of fault tolerance.The first part of this thesis proposes a state of the art of dependability, at first in a generic way before being specified in the context of control architectures. The second part allows us to detail the proposed methodology to identify potential errors of a robot and respond using the means of fault tolerance. The third part presents the experimental context and application in which the proposed methodology will be implemented and described in the fourth part of this manuscript. An experiment highlighting specific aspects of the methodology is detailed in the last part.

This master thesis describes the design and implementation of control sys- tem for autonomous robot which is able to run through user defined points in unknown environment without colliding with obstacles. The work contains analysis of the available hardware and software solutions, modular design with control system implementation divided into separate subsystems (control, lo- calization, route planning, driving the robot using Hermit curves and low-level hardware control). The work also contains explanation of rework of the school robotic platform. The implementation was tested on a created robotic platform. Driving the robot along the Hermit curve allows smooth and in some cases quicker passage through defined points, than passage consisting of rotations on the spot and direct movements. 1

The Ph.D. thesis is focused on using the reinforcement learning for four legged robot control. The main aim is to create an adaptive control system of the walking robot, which will be able to plan the walking gait through Q-learning algorithm. This aim is achieved using the design of the complex three layered architecture, which is based on the DEDS paradigm. The small set of elementary reactive behaviors forms the basis of proposed solution. The set of composite control laws is designed using simultaneous activations of these behaviors. Both types of controllers are able to operate on the plain terrain as well as on the rugged one. The model of all possible behaviors, that can be achieved using activations of mentioned controllers, is designed using an appropriate discretization of the continuous state space. This model is used by the Q-learning algorithm for finding the optimal strategies of robot control. The capabilities of the control unit are shown on solving three complex tasks: rotation of the robot, walking of the robot in the straight line and the walking on the inclined plane. These tasks are solved using the spatial dynamic simulations of the four legged robot with three degrees of freedom on each leg. Resulting walking gaits are evaluated using the quantitative standardized indicators. The video files, which show acting of elementary and composite controllers as well as the resulting walking gaits of the robot, are integral part of this thesis.

Cette recherche, qui s'inscrit dans le cadre des travaux en didactique de l'informatique de Baron et Bruillard, analyse la façon dont des enseignants du primaire, non experts en informatique,conçoivent et mettent en œuvre des scénarios impliquant des robots pédagogiques de sol dans leurs classes. La mise en œuvre de ces robots a été étudiée avec l'objectif, d'apporter un éclairage sur leurs possibles apports pédagogiques. Il s'est agi de préciser comment ces enseignants parviennent à définir des situations didactiques de ces objets de connaissances auxquels ils n'ont jamais été confrontés et d'analyser la manière dont ils parviennent à développer chez les élèves une pensée informatique en actes. L'activité des élèves a été analysée, au travers l'approche instrumentale (Rabardel), en vue de comprendre de quelle manière l'apprentissage de concepts en informatique émerge de ces activités. La question des apprentissages des concepts et méthodes propres au domaine informatique par le biais de la robotique est analysée en prenant appui sur la théorie des champs conceptuels (Vergnaud). Cette recherche apporte des éléments permettant de comprendre comment ces enseignants parviennent, de manière intuitive, à développer et à mettre en œuvre des scénarios pour enseigner quelques concepts informatiques. Elle témoigne de leur capacité à intégrer des objets tangibles ou symboliques dans des séances d'apprentissage en informatique, en procédant à une analyse préalable à minima du fonctionnement du robot. Au cours des activités dans lesquelles ils sont mobilisés, outils robotiques et aides pédagogiques accompagnent les apprentissages. Sur le plan des apprentissages, les élèves se sont forgés, par des démarches d'instrumentation et d'instrumentalisation, des instruments et des méthodes pour comprendre l'objet informatique. Les concepts et notions en jeu sont particulièrement dépendants des contextes technologiques spécifiques à chacun des robots. Des méthodes propres à la production logicielle ont permis le séquençage de l'activité de programmation en phases de spécification, conception, réalisation et mise au point. Des paradigmes de programmation ont aussi été approchés, comme la programmation procédurale dans le cas du robot Bee-Bot et la programmation événementielle dans le cas de l'étude du comportement du robot Thymio. En dehors du domaine informatique, la résolution de problème, en étant placée au cœur des scénarios, a permis aux élèves de développer des démarches de tâtonnements, d'essais-erreurs dans un contexte de travail en petit groupe favorisant les échanges et les interactions entre les élèves. / This 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.