Search

International audience; This paper formalizes, under a single common Multi-Sensor-Based Predictive Control framework, five different types of parking maneuvers: perpendicular, diagonal for both forward and backward motions and parallel for backward motions. Since, from a practical point of view, forward parallel parking is usually not advisable, it is not addressed in this work. By moving the effort from motion planning to control, the parking tasks can be completely defined solely from the detected empty parking spots. Additionally, the classical compromise between completeness and computational efficiency when compared to exploration-based path planning techniques is eliminated. The results of a few individual cases are presented and compared against a state of the art path planning approach to illustrate the behavior and performance of the proposed framework as well as results from exhaustive simulations to assess its convergence. As shown in the convergence analyses, the presented approach allows to park from virtually any sensible initial pose. Finally, real experimentation using a robotized Renault ZOE shows the validity and robustness in the convergence domain of the presented approach.

This thesis is part of a more general research effort aimed at contributing to the fundamental study of multifunctional wear phenomenons, which some stainless steel components are subjected to in nuclear reactors. In reactor, the steel’s surface is subjected to both water radiolysis, caused by irradiation, and periodical friction damaging it’s protective oxide layer. Wear phenomenons in reactors are well controlled, but a better understanding of synergetic effects between those processes remain important. To this end, these processes need to be decorrelated in a controlled manner and within a simplified system. This thesis is the first step in this research effort. An experimental cell, recently developed in cooperation between the group CorrIS/MatéIS of the INSA Lyon and the group ACE of the IP2I UCBL Lyon, was used. Mounted at the end of a charged particles accelerator, it allow to generate radiolysis at the interface between a Na2SO4 2x10-2 mol.L-1 solution and a 316L stainless steel sample. This cell is equipped with a pin generating friction at the surface of the sample while recording the force and torque applied. Radiolysis of the solution is achieved using protons irradiation. Protons goes through the steel sample and stops in the solution. Experiments were carried out using either the cyclotron of the CEMHTI (Orléans, France) with 16 MeV protons or using the Van de Graaff accelerator of the IP2I (Lyon, France) with 2-3 MeV protons. The effects of radiolysis at the interface were investigated in situ using a three electrodes setup. Various electrochemical techniques were used. Among those Electrochemical Impedance Spectroscopy (EIS) was used to characterize the protective oxide layer formed on the surface of stainless steel. After devising a protocol of preparation, ensuring a repeatable initial surface state of the sample, the effects of radiolysis at the interface were studied. Effects of the protons flux, as well as the distance between the interface and the Bragg peak, on the open circuit potential (OCP) were studied. Simulation of radiolysis at the interface were performed using the SRIM and CHEMSIMUL softwares, then the order of magnitude of H2O2 concentration created by irradiation was determined through dosages. An influence of the irradiation on the corrosion potential of steel as well as on the current density measured were found. EIS measurements have shown an effect of the irradiation on the migration kinetics of species through the protective oxide layer. Reproducing the effects of irradiation on the corrosion potential and current density was possible by adding H2O2 at a concentration two orders of magnitude higher than one measured after irradiation by dosage. Migration kinetics under irradiation could however not be reproduced by H2O2 addition, whatever the added concentration. Finally, electrochemical noises measurements were performed allowing to quantify the reformation kinetics under irradiation of the oxide layer after its removal by friction. By modulating the energy of the protons, it was shown that the corrosion potential reached was independent of the distance between the Bragg peak and the interface. Moreover, EIS measurements shows modifications of the migrations kinetics through the passive layer under irradiation. This set of results lead to the hypothesis that the radicals produced closest to the interface are the one controlling the electrochemical state of the interface. It is proposed that water molecules present inside hydrated layers of the oxide (hydroxides) are subject to radiolysis, thus greatly reducing the migration distance of species in the oxide layer, Cette thèse s’inscrit dans le cadre plus général de travaux de recherche ayant pour but de contribuer à l’étude fondamentale de phénomènes d’usure multifonctionnels que subissent certaines pièces en acier inoxydables dans les réacteurs nucléaires. En réacteur, la surface de l’acier est en effet soumise à la radiolyse de l’eau générée par l’irradiation ainsi qu’à des frottements périodiques, conduisant à l’endommagement de la couche d’oxyde protectrice. Les phénomènes d’usure en réacteurs sont bien maîtrisés, mais il reste important de mieux comprendre les mécanismes synergiques entre ces différents processus et donc de les décorréler de manière contrôlée dans un système simplifié. Cette thèse constitue une première étape dans ce programme de recherche. Un dispositif expérimental récemment développé dans le cadre d’une collaboration entre le groupe CorrIS/MatéIS de l’INSA de Lyon et le groupe ACE de l’IP2I à l’UCBL, a été utilisé. Installé en ligne sur un accélérateur de particules chargées, il permet de générer la radiolyse à l’interface entre une solution de Na2SO4 2x10-2 mol.L-1 et un échantillon d’acier 316L . Un pion permet également de frotter la surface de l’acier et de mesurer la force et le couple appliqués. Les radicaux sont produits à l’interface par un faisceau de protons qui traversent la feuille d’acier avant de s’arrêter dans la solution. Les expériences ont été réalisées soit sur le cyclotron du CEMHTI (Orléans France) avec des protons de 16 MeV soit sur l’accélérateur Van de Graaff 4MV de l’IP2I (Lyon France) avec des protons de 2-3 MeV. Les effets de la radiolyse à l’interface ont été suivis in situ à l’aide d’un montage à trois électrodes. Diverses techniques électrochimiques ont ainsi été mises en œuvre, dont la Spectroscopie d’Impédance Electrochimique (SIE) qui permet de caractériser le film d’oxyde à la surface du métal. Après la mise au point d’un protocole de préparation de la surface de l’échantillon, garantissant la reproductibilité de l’état initial, les effets de la radiolyse à l’interface ont été étudiés. L’influence du flux de protons et de la distance pic de Bragg-interface sur le potentiel libre de l’électrode d’acier a été étudiée. Des simulations de la radiolyse à l’interface ont été réalisées via les logiciels SRIM et CHEMSIMUL, puis l’ordre de grandeur de la concentration d’H2O2 créée par l’irradiation a été quantifiée. L’influence de l’irradiation sur les potentiels de corrosion ainsi que sur les densités de courant mesurées, a été mise en évidence. Les mesures de SIE ont également permis de montrer le rôle de l’irradiation sur les processus de migration des espèces à l’interface film passif-solution. Il a été montré que les effets de l’irradiation sur le potentiel et les densités de courants peuvent être simulées par ajout d’H2O2. La concentration d’H2O2 nécessaire est toutefois supérieure de deux ordres de grandeurs à celle mesurée après irradiation. Les processus de migration sous irradiation ne peuvent pas être reproduits par ajout d’H2O2, quelle que soit la concentration. Enfin, des mesures de bruits électrochimiques, ont permis de quantifier les cinétiques de reformation, sous irradiation, de l’oxyde protecteur de surface après son abrasion par frottement. En faisant varier l’énergie des protons incidents, il a été montré que la valeur du potentiel est indépendante de la distance du pic de Bragg à l’interface. De plus, la SIE montre une modification des processus de migration des espèces au travers du film passif sous irradiation. Cet ensemble de résultats conduit à émettre l’hypothèse que ce sont les radicaux produits au plus près de l’interface qui contrôlent l’état électrochimique de l’interface. Il est proposé que ce sont les molécules d’eau présentes dans les strates hydratées de l’oxyde (hydroxydes) qui sont radiolysées, réduisant ainsi fortement les distances de migration des espèces dans l’oxyde

[Departement_IRSTEA]Ecotechnologies [TR1_IRSTEA]INSPIRE; International audience; This paper addresses the automation of farm vehicle manoeuvres in headlands. A trajectory generation strategy is presented to plan reverse turns. Then, both steering and speed control algorithms are presented to guide the vehicle on the planned path, based on its kinematic model extended with additional sliding parameters and on predictive approaches. Real world experiments have been carried out on a low adherent terrain with an experimental mobile robot. At the end of each row, the sequences of the reverse turn are automatically generated to connect the next track to be followed, and the manoeuvre is autonomously performed by the vehicle. Reported experimental results demonstrate the capabilities of the proposed algorithms in full scale conditions.

International audience; In this paper, we present a generic framework for urban vehicle navigation using a topological map. This map is built by taking into account the non-holonomic behaviour of the vehicle. After a localization step, a sensory route is extracted to reach a goal. This route is followed using a sensor-based control strategy, based on the vehicle model and computed from the state extracted from the current and the desired sensory images. In that aim, a generic model is proposed for visual sensors. Experiments with an urban electric vehicle navigating in an outdoor environment have been carried out with a fisheye camera using a single camera and natural landmarks. A navigation along a 1700-meter-long trajectory validates our approach.

National audience

International audience; Robotic manipulation of everyday objects and dependable execution of household chores is one of the most desired and challenging skills for future service robots. Most of the current research in robotic grasping is limited to pick-and-place tasks, without paying attention to the whole range of different tasks needed in human environments, such as opening doors, interacting with furniture, household electrical appliances , etc. In this article, a sensor-based integrated framework for specifying both the grasp and the task is presented, with the goal of motivating dependable task-oriented grasping and manipulation. The grasp is defined as a desired task-suitable relationship between the robot hand and the object being manipulated. The task is defined under the task frame formalism [1], which allows to specify tasks for sensor-guided dependable interaction. Some guidelines for sensor-based execution of tasks defined under the proposed formalism are also given. Three different examples of dependable manipulation tasks, performed by three different robots, are presented, making use of the proposed approach and disparate sensor information: book grasping by tactile-force integration, door opening by vision-force control, and force-guided humanoid interaction with kitchen furniture.

International audience

International audience; The use of centrifugal spreaders for application of granular fertilizers raises concern about application accuracy. These spreaders enable to distribute almost uniform deposits with regularly spaced parallel tramlines but lead over and under fertilizer application when distances separating the successive tractor trajectories are not constant or if paths are not parallel between them. The aim of this study is to propose an optimization-based method to compute optimal variables for uniform fertilizer application. Mechanical constraints are also introduced so that the calculated parameters can be afterwards used as reference variables for the control of the spreader. The resulting improvements are illustrated by numerical examples for a field with geometrical singularities.

[Departement_IRSTEA]EA [TR1_IRSTEA]TESAD / TECHPROPRE; We study an approach for the optimization of spreading performed by centrifugal spreaders in order to minimize environmental effects due to application errors. Faced with a large scale problem, we divide the domain of study into subdomains so that each tramline is individually dealt with. In order to take into account the mechanical limits of the device, some inequality constraints are introduced. After cost function discretization, we use an augmented lagrangian algorithm associated with a L-BFGS technique to solve the problem. Results are presented for parallel and non parallel paths.

International audience; In this paper, the problem of autonomous navigation is addressed starting from the use of RTK-GPS until the use of vision as the main sensor. Using a decoupling strategy, it is possible to control separately lateral and longitudinal control in trajectory following tasks. Recent advances in vision, make possible to localize an urban vehicle in regard with a previously recorded trajectory. The concept of sensory memory (GPS, Vision), which tra-duces the learnt trajectory, is introduced and used to guide autonomously the vehicle. Then, an extension to vehicle platoon is described. Different kind of longitudinal control strategies are discussed : from a near to near approach, to a global strategy. Finally, in order to manage different kind of scenario which occur in a platoon context, a monitoring module is developed. All theses aspects are currently addressed through the BODEGA (ROBEA-CNRS national interdisciplinary research program) and MOBIVIP (PREDIT3 national research program) projects.

International audience; This paper describes and illustrates the theoretical framework of an original obstacle avoidance module which aims to assist the robot guidance while navigating into a cluttered environment. This module, called CAM (Control Assistance Module) corrects the control inputs from a navigation module according to local observations, extracted from range data given by embedded sensors. It aims to warrant the robot security but keeps on taking into account the control orders from the navigation module, in order to satisfy the navigation task for the best. The obstacle avoidance approach is inspired by a reflex avoidance method, which consists on preserving a virtual zone around the robot. In an original way, when an obstacle enters into this virtual zone, the CAM plans dynamicly a judicious path to be followed to keep the zone intact. Avoidance control orders are then computed and merged with navigation orders by the CAM, according to a simple rule which privileges the robot's security. The CAM has been implemented and validated on a cart-like mobile robot, equipped with a sonar sensors belt. It also has been put on the test when the robot was teleoperated into an unknown environment for exploration.

[Departement_IRSTEA]EA [TR1_IRSTEA]TESAD / MOST; National audience; Si la problématique du suivi de trajectoire de robots mobiles sous hypothèse de roulement sans glissement des roues sur le sol est aujourd'hui satisfaisante dans certaines conditions favorables (pas d'annulation de la vitesse, bonnes conditions d'adhérence, ...), il n'en va pas de même pour les véhicules évoluant en milieu naturel. En effet, les faibles propriétés d'adhérence des terrains d'évolution ne permettent pas une transposition directe des algorithmes de suivi, menant alors à des pertes importantes de précision lorsque l'hypothèse de roulement sans glissement est violée. En outre les propriétés géométriques et dynamiques des engins évoluant sur de tels terrains (inertie, garde au sol, ...) ainsi que les actionneurs employés génèrent des phénomènes de retard augmentant lors de phases transitoires, les erreurs de suivi enregistrées. Dans cet article plusieurs développements théoriques (basés sur des principes d'observateur, de commande adaptative et prédictive), visant à conserver le même niveau de précision que celui enregistré dans le cas des véhicules routiers, sont étudiés et testés en vraie grandeur sur des engins agricoles.

National audience; Cet article présente une nouvelle approche de la com-mande référencée vision des mécanismes parallèles. Elle est basée sur le concept de redondance métrologique qui simplifie leurs modèles géométriques en utilisant des cap-teurs proprioceptifs additionnels. En observant les jambes du mécanisme, la vision fournit, dans un repère cartésien, une mesure extéroceptive de l'état interne du mécanisme. Formellement, il s'agit d'exprimer, par un concept original , la commande cinématique référencée vision des mécanismes parallèles. La commande utilise la mesure vi-suelle de la direction des jambes, plutôt qu'une mire placée sur l'organe terminal. La méthode est illustrée et validée par une simulation de la plateforme de Gough-Stewart. Abstract This paper presents a novel approach for vision-based control of the end-effector of parallel mechanisms. It is based on the metrological redundancy paradigm, which simplifies their kinematic models by introducing additional proprioceptive sensors. By observing the mechanism legs, vision replaces advantageously these sensors by delivering , in a Cartesian frame, an exteroceptive measurement of the internal state of the mechanism. Formally, the latter is expressed by an original concept of vision-based kinematics for parallel mechanisms. Based on it, control is derived that visually servoes the direction of the legs, rather than the end-effector pose. The method is illustrated and validated on a Gough-Stewart platform simulation.

National audience; Pour résoudre les problèmes liés au trafic important dans les métropoles, de nouvelles alternatives , appelés "Urban Transportation Systems", voient le jour. Celles-ci sont basées sur des véhiculeś electriques en libre accès. Une fonctionnalité nécessaire de ces systèmes est leur capacitéà se déplacer en convoi. Dans ce papier, est présenté le convoi de ces véhicules automatisés, s'appuyant sur des capteurs RTK-GPS et une communication inter-véhicules. La commande développée est basée sur une stratégie de commande globale ; en fait, l'état du convoi est pris en compte. En fait, le com-portement du convoi estétudiéà travers lesétats du leader et du véhicule situé immédiatement devant. La distance inter-véhicules est prise comme la distance curviligne le long d'une trajectoire de référence. Se basant sur la théorie de commande non-linéaire, les contrôles latéral et longitudinal sont traités indépendamment. Afin de prodiguer un confort aux passagers, des fonctions de supervision sont greffées au système de contrôle. Puis, des expérimentations, portées sur véhicules réelles, et des simulations de longs convois sont présentées.

International audience; Satisfactory results of automatic guidance of farm vehicles have been achieved in previous works as long as vehicles move without sliding, but unfortunately in agricultural applications, sliding always occurs inevitably which causes loss of accuracy. In this paper the problem of path following control of autonomous farm vehicles subject to sliding is addressed. To take sliding effects into account, a vehicle-oriented kinematic model is built in which sliding effects are integrated in the form of additive disturbances to the ideal kinematic model. By transforming the vehicle-oriented kinematic model into a perturbed chained system, a sliding mode controller, which is robust not only to the sliding effects but also to the input noise, is designed with the help of the natural algebraic structure of chained systems. Simulation results show that the proposed control law can guarantee high path-following accuracy even in the presence of sliding.

Since Global Navigation Satellite Systems are able to supply very accurate coordinates of a point (about 2 cm with a RTK GPS), such a sensor is very suitable to design vehicle guidance system. It is especially the case in agricultural tasks where a cantimeter precision is often required (seeding, spraying,...). To answer to growing high precision agriculture principle demand, several control laws for automated vehicle guidance systems are able to supply an acceptable steering accuracy as long as vehicle does not slide (path tracking on even ground with good adherence properties...), what alas inevitably occurs in agricultural tasks. Several principles are here presented to steer vehicle whatever properties of ground and path to be followed are. In this paper two extended kinematic models with sliding accounted are presented. They allow to describe vehicle dynamics in all guidance conditions. Via these models, new non linear control laws can be designed (depending on model structures), which integrate sliding effects. Their capabilities are investigated through experimental tests., Depuis que les systèmes de navigation par satellites peuvent fournir des coordonnées très précises d'un point (environ 2cm avec un GPS RTK), de tels capteurs sont très utilisés pour concevoir les système de guidage de véhicule. C'est particulièrement le cas dans le domaine agricole où une précision centimètrique est souvent exigée (semis, pulvérisation...). Pour répondre à la demande croissante de cette agriculture de précision, plusieurs lois de commande ont été développées pour le guidage automatique de véhicules basées sur ce capteur GPS. De tels systèmes peuvent assurer une précision de guidage acceptable dés lors que le véhicule ne glisse pas (suivi de trajectoire sur sol plat avec de bonnes propriétés d'adhérence...), ce qui hélas se produit inévitablement dans les travaux agricoles. Plusieurs principes sont ici présentés pour guider le véhicule quelques soient les propriétés du sol et le chemin à suivre. Dans cet article, deux modèles cinématiques étendus avec prise en compte du glissement sont présentés. Ils permettent de décrire les dynamiques du véhicule dans toutes les conditions de guidage. Par l'intermédiaire de ces modèles, de nouvelles lois non linéaires de commande peuvent être conçues, qui intègrent les effets du glissement. Leurs performances sont étudiées à travers d`essais expérimentaux.

International audience; We deal in this paper with an original visual servoing method intended to control a nonholonomic wheeled mobile robot. These works are a part of a complete image-based framework for navigation of a mobile robot, using only a monocular vision system embedded on the robot. The navigation system uses a set of relay images to enable the robot to reach a goal configuration expressed as a desired attitude of visual landmarks in the image plane. This paper particularly focuses on a suitable image-based control law of a nonholonomic robot for navigating in image database, with the assumption that the ground is obstacle free. We show how computing a suitable homographic transformation between the current camera image and a desired one allows to estimate an error vector on the state of the robot relative to a virtual straight line. This line is joined and followed thanks to a control law designed according to a well suited chained system of the state of the robot. Simulations illustrate this point, and experimental results for a two wheels driven robot confirm the validity of our approach.

International audience; In the last decades, robotics has exerted an important role in the research on diverse knowledge domains, such as, artificial intelligence, biology, neuroscience and psychology. In particular, the study of knowledge representation and thinking, has led to the proposal of cognitive architectures; capturing essential structures and processes of cognition and behavior. Robotists have also attempted to design automatic systems using these proposals. Though, certain difficulties have been reported for obtaining efficient low-level processing while sensing or controlling the robot. The main challenges involve the treatment of the differences between the computational paradigms employed by the cognitive and the robotic architectures. The objective of this work, is to propose a methodology for designing robotic systems capable of decision making and learning when executing manipulative tasks. The development of a system called the Cognitive Reaching Robot (CRR) will be reported. CRR combines the advantages of using a psychologically-oriented cognitive architecture, with efficient low-level behavior implementations through the visual servoing control technique.

International audience; One of the main drawbacks of vision-based control that remains unsolved is the poor dynamic performances caused by the low acquisition frequency of the vision systems and the time latency due to processing. We propose in this paper to face the challenge of designing a high-performance dynamic visual servo control scheme. Two versatile control laws are developed in this paper: a position-based dynamic visual servoing and an image-based dynamic visual servoing. Both control laws are designed to compute the control torques exclusively from a sequential acquisition of regions of interest containing the visual features to achieve an accurate trajectory tracking. The presented experiments on vision-based dynamic control of a high-speed parallel robot show that the proposed control schemes can perform better than joint-based computed torque control.

National audience; This paper is about a Multi-Agent based solution to control and coordinate team-working mobile robots moving in unstructured environments. Two main contributions are considered in our approach. The rst contribution of this paper is about the Multi-Agents System to Control and Coordinate teAmworking Robots (MAS2CAR) architecture, a new architecture to control a group of coordinated autonomous robots in unstructured environments. MAS2CAR covers three main layers: (i) the Physical Layer (ii) the Control Layer and (iii) the Coordination Layer. The second contribution of this paper is about the multi-agent system (MAS) organisational models aiming to solve the key cooperation issues in the coordination layer, the software components designed based on Utopia a MAS framework which automatically build software agents, thanks to a multi-agent based organisational model called MoiseInst . We provide simulation results that exhibit robotics cooperative behavior related to our scenario, such as multi-robots navigation in presence of obstacles (including trajectory planning, and reactive aspects) via a hybrid control.

International audience; This paper addresses the problem of controlling the dynamics of parallel kinematic manipulators from a global point of view, where modeling, sensing and control are considered simultaneously. The methodology is presented through the examples of the Gough-Stewart manipulator and the Quattro robot.