Suchergebnisse

As the capabilities of the mobile robots as well as their abilities to perform more tasks in an autonomous manner are increased, we need to think about the interactions that humans will have with these robots. Human-robot interaction (HRI) has recently received considerable attention in the academic community, government labs, technology companies, and through the media. The interdisciplinary nature of HRI requires researchers in the field to understand their research within a broader context. Since natural language is the easiest and most natural mode of communication for humans, it is desirable to use it to instruct the robot and to generate easy-to-understand messages for the user. Using natural language to teach a navigation task to a robot is an application of a more general instruction-based learning methodology. It can be used to instruct the robot with higher-level goals or to handle certain behaviors and modify their execution. One effective way is to describe the route to the robot in a multimodal way. On the other hand, significant progress has been made towards stable robotic bipedal walking in the last few years. This is creating an increased research interest in developing autonomous navigation strategies which are tailored specifically to humanoid robots. Efficient approaches to perception and motion planning, which are suited to the unique characteristics of bipedal humanoid robots and their typical operating environments, are receiving special interest. One important area of research involves the design of algorithms to compute robust navigation strategies for humanoid robots in human environments. Therefore, autonomous robot navigation based on route instruction is becoming an increasingly important research topic with regard to both humanoid and other mobile robots. In this dissertation, the problem of humanoid robot navigation in indoor environments is addressed. A complete framework is presented for humanoid robot navigation based on a multimodal cognitive interface. First, a spatial language to describe route-based navigation tasks for a mobile robot is proposed. This language is implemented to present an intuitive interface that enables novice users to easily and naturally describe a route to a mobile robot in indoor environments. An instruction interpreter is implemented to analyze the user's route to generate its equivalent symbolic and topological map representations which are used as an initial path estimation for the humanoid robot. Second, a robust lightweight object processing system with a high detection rate is developed. It can actually be used by mobile robots and meet their hard constraints to recognize landmarks during navigation. A landmark processing system is developed to detect, identify, and localize different types of landmarks during robot navigation in indoor or miniature city environments. The system is based on a two-step classification stage which is robust and invariant towards scaling and translations. By combining the strengths of appearance-based and model-based object classification techniques, it provides a good balance between fast processing time and high detection accuracy. Finally, a time-efficient hybrid motion planning system for a humanoid robot in indoor environments is implemented. The proposed technique is a combination of sampling-based planner and D* Lite search to generate dynamic footstep placements in unknown environments. A modified cylinder model is used to approximate the trajectory for the robot's body-center during navigation. It calculates the actual distances required to execute different actions of the robot and compares them to the distances from the nearest obstacles. D* Lite search is then used to find dynamic and low-cost footstep placements within the resulting configuration space. ; Da die Fähigkeiten von mobilen Robotern einschließlich ihrer Möglichkeiten, Aufgaben autonom durchzuführen, erweitert wurden, muss die Interaktion zwischen Mensch und Roboter neu betrachtet werden. Human-Robot-Interaction (HRI) ist ein aktuelles Thema in der Forschung, in Technologie-Unternehmen und in den Medien. Der interdisziplinäre Charakter des HRI-Bereiches erfordert Forschung innerhalb eines breiten Themenkomplexes. Da natürliche Sprache das einfachste und natürlichste Mittel der Kommunikation für Menschen ist, ist es wünschenswert, diese Form der Kommunikation auch bei der HRI zu nutzen, um einem Roboter Anweisungen zu geben und leicht verständliche Botschaften für den Benutzer zu generieren. Die Verwendung natürlicher Sprache zur Instruierung bei Navigations-Aufgaben ist eine Anwendung einer allgemeineren instruktions-basierten Lernmethodologie. Dem Roboter können so übergeordnete Ziele mitgeteilt werden, bestimmte Verhaltensweisen geändert oder auch die Ausführung einzelner Aktionen modifiziert werden. Eine effiziente Methode zur Beschreibung der Route ist die Verwendung multimodaler Anweisungen. Weil die vergangenen Jahre einen bedeutenden Fortschritt auf dem Gebiet der humanoiden Roboter und des stabilen zweibeinigen Gehens gebracht haben, besteht ein verstärktes Forschungsinteresse an der Entwicklung autonomer Navigationsstrategien, die speziell auf humanoide Roboter zugeschnitten sind. Von besonderem Interesse sind effiziente Ansätze zur kombinierten Perzeptions- und Aktionsplanung, die an die speziellen Eigenschaften von zweibeinigen humanoiden Robotern und ihre typischen Betriebsumgebungen angepasst sind. Ein wichtiges Gebiet der Forschung ist der Entwurf von Algorithmen zur Berechnung von robusten Navigations-Strategien für humanoide Roboter in menschlicher Umgebung. Aus diesem Grunde ist die auf Routen-Instruktion beruhende autonome Roboter-Navigation ein zunehmend interessantes Thema im Hinblick auf humanoide und andere mobile Roboter. Diese Dissertation befasst sich mit dem Problem der humanoiden Roboter-Navigation in Innenräumen. Es wird ein komplettes Framework für humanoide Roboter-Navigation basierend auf einer multimodalen Schnittstelle vorgestellt. Zunächst wird eine formale Sprache eingeführt, mit der die routen-basierten Navigationsaufgaben beschrieben werden können. Diese Sprache stellt eine intuitive Schnittstelle bereit, mit der auch unerfahrene Anwender leicht einen mobilen Roboter in einer Route in Innenräumen instruieren können. Ein Befehls-Interpreter analysiert die Benutzer-Eingabe und generiert entsprechende symbolische und topologische Darstellungen, die als erste Pfad-Schätzung für den humanoiden Roboter verwendet werden. Des Weiteren wird in dieser Arbeit ein robustes und effizientes Objekterkennungssystem mit einer hohen Erkennungsrate entwickelt. Es kann bei mobilen Robotern eingesetzt werden und erfüllt die Anforderung, Landmarken während der Navigation zu erkennen. Das Landmarken-Detektions-System ist in der Lage, während der Roboter-Navigation in einer Miniatur-Stadt verschiedene Typen von Landmarken zu detektieren, identifizieren und zu lokalisieren. Das System basiert auf einem zweistufigen Klassifikations-Prozess, der robust und invariant gegenüber Skalierung und Translation ist. Durch die Kombination der Stärken der erscheinungs-basierten und modell-basierten Objekt-Klassifikation bietet es einen guten Kompromiss zwischen schnellen Bearbeitungszeiten und hoher Erkennungsgenauigkeit. Ebenfalls Bestandteil dieser Arbeit ist die Implementierung eines zeiteffizienten hybriden Bewegungs-Planungs-Systems für humanoide Roboter in einer Innenraum-Umgebung. Die vorgeschlagene Technik ist eine Kombination aus Sampling-basierter Planung und "D * Lite"-Suche, die ermöglicht, dynamisch Tritt-Platzierungen in unbekannten Umgebungen zu erzeugen. Ein modifiziertes Zylinder-Modell wird verwendet, um die Trajektorie des Roboters während der Navigation näherungsweise zu bestimmen. Die Planungskomponente berechnet die benötigten Freiräume, um verschiedene Aktionen des Roboters auszuführen und vergleicht sie mit der aktuellen Entfernung zu den nächstgelegenen Hindernissen. "D* Lite"-Suche wird dann verwendet, um eine dynamische und effiziente Platzierung der Schritte innerhalb des resultierenden Konfigurations-Raumes zu finden.

Searching and tracking are important behaviours for a mobile service robot to assist people, to search-and-rescue and, in general, to locate mobile objects, animals or humans. Even though searching might be evident for humans, for robots it is not, since it requires exploring, handling noisy sensors, coping with dynamic obstacles, and coordination in the case of multiple agents. In this thesis, we present several methods to search and track a person in an urban environment. All methods were fist tested extensively in simulation and then in real-life, using one or two mobile service robots, called Tibi and Dabo. The robots have laser range finders, which are used to navigate, to detect obstacles and to detect people's legs. Since we focus on search-and-track methods, we use existing methods for robot navigation, for people detection and person recognition. First tests are done with the hide-and-seek problem, in which the robot learns to catch the hider. Concretely, a Mixed Observable Markov Decision Process (MOMDP) model is used, in which the seeker's location is fully observable and the hider's location partially observable. Since the computational complexity depends on the number of states, we propose a hierarchical on-line method that reduces the state space by grouping them together. Although the method worked properly in simulation, in the real-life experiments the results were not satisfying and the on-line policy calculation was not fast enough to work in real-time. To handle larger environments, work in continuous state space and run in real-time, we propose to use an approach, the Continuous Real-time POMCP (CR-POMCP), that does Monte-Carlo simulations to learn a policy. The method performed correctly in simulation, but on the real robot it resulted in slow zigzag movements. Therefore, a new method is proposed, which uses the highest probable locations, according to its probability map (belief). Since the belief propagation of the POMCP resembles how a Particle Filter (PF) works, we also propose a method that uses a PF to maintain the belief. The PF method has to handle lack of observations, therefore, we introduce a special weight function. Both belief update methods take into account sensor and actuator noise, false negative detections, false positive detections (for a short time) and dynamic obstacles. Finally, a cooperative distributed multi-agent method is presented, it makes use of the previous belief update functions, but it uses all the agents' observations. Next, the search locations are assigned by exploring the working environment, taking into account: the belief, the distance to the search location and if another agent already will search close to it. Summarizing, the main contributions of this thesis are several methods to search and track a person in an urban environment with one or more mobile service robots. All these methods have been shown to work through a set of simulations and real-life experiments. ; La cerca i el seguiment de persones són comportaments importants per un robot mòbil de servei per poder assistir, trobar i ajudar als humans, i en general, per localitzar objectes, animals o vianants. Tot i que la cerca és fàcil per als humans, no ho és per a un robot, ja que requereix exploració, maneig de soroll de sensors, fer front als obstacles dinàmics, i la coordinació en el cas de múltiples agents. En aquesta tesi presentem diferents mètodes per a buscar i seguir a una persona en un entorn urbà. Tots els mètodes han estat provats extensivament en simulació i després en el món real, utilitzant dos robots mòbils de servei, la Tibi i en Dabo. Els robots utilitzen sensors làser per a navegar, detectar obstacles i detectar les cames de les persones. Atès que aquest treball es centra en mètodes de cerca i seguiment, s'han usat els mètodes existents per a la navegació del robot, la detecció i el reconeixement de persones. Primerament, s'han fet proves amb el conegut joc del fet i amagar, on el robot aprèn a trobar l'amagador. S'ha fet servir el model Mixed Observable Markov Decision Process (MOMDP), on la posició del trobador és completament visible i la posició de l'amagador és parcialment visible. Degut a que la complexitat computacional depèn del nombre d'estats, es proposa un mètode jeràrquic en línia que redueix l'espai d'estats, tot agrupant-los. Tot i que el mètode va funcionar correctament en simulació, en els experiments reals els resultats no van ser satisfactoris, i el càlcul de la política no va ser prou ràpid com per treballar en temps real. Per tal de fer front a entorns de més superfície, treballar en l'espai continu i executar en temps real, proposem un nou enfocament, el Continuous Real-time POMCP (CRPOMCP), que fa simulacions de Monte-Carlo per aprendre una política. El mètode va funcionar correctament en l'entorn simulat, però a l'entorn real el robot realitzava lents moviments en zig-zag. Per tant, es proposa un mètode nou, que utilitza els llocs amb més alta probabilitat, d'acord amb el seu mapa de probabilitats (belief). Atès que la propagació de les probabilitats en el POMCP és similar al funcionament d'un filtre de partícules (PF), proposem, a més, un mètode que utilitza un PF per mantenir el belief. El mètode de PF ha de manejar la manca d'observacions. Per tant, introduïm una funció del pes especial. Tots dos mètodes de creences tenen en compte el soroll dels sensors i actuadors, la detecció de falsos negatius i positius (per a un curt període de temps) i els obstacles dinàmics. Finalment, es presenta un mètode multi-agent distribuït cooperatiu, que fa ús de les anteriors funcions d'actualització de la creença (belief), i a més utilitza totes les observacions dels agents. En el proper pas, les ubicacions de cerca s'assignen mitjançant l'exploració de l'entorn de treball, tenint en compte la creença, la distància a la ubicació de cerca i si un altre agent ja buscarà a prop d'ella. En resum, les principals contribucions d'aquesta tesi són diversos mètodes per a la cerca i seguiment d'una persona en un entorn urbà amb un o més robots de serveis mòbils. Tots aquests mètodes han demostrat que funcionen a través d'un conjunt de simulacions i experiments en la entorn real dinàmics. ; La búsqueda y el seguimiento de personas son comportamientos importantes para un robot móvil de servicio para poder asistir, buscar y ayudar a la gente, y en general, para localizar un objeto, animal o humano. Aunque la búsqueda puede parecer muy fácil para los humanos, para los robots no lo es, ya que requiere explorar, manejar ruido de sensores, enfrentarse con obstáculos dinámicos y la coordinación en el caso de haber más agentes. En esta tesis, presentamos diferentes métodos para buscar y seguir a una persona en un entorno urbano. Todos los métodos han sido probados excesivamente en simulación y en experimentos reales, usando uno o dos robots móviles de servicio, Tibi y Dabo. Los robots tienen localizadores láser, los cuales se usan para navegar, detectar obstáculos y detectar piernas. Ya que el principal enfoque en este trabajo son los métodes de buscar-y-seguir, utilizamos métodos existentes para la navigación del robot, la detección de personas y el reconecimiento del humano. Las primeras pruebas se hicieron con el juego de escondite, en el cual el robot aprende a buscar el ocultador. Concretamente, usamos un modelo MOMDP, donde la posición del buscador es completamente observable y la posición del ocultador lo es parcialmente. Como la complejidad computacional depende del número de estados, proponemos un método en línea jerárquica, que reduce el espacio de los estados agrupándolos. Aunque el método funcionó correctamente en simulación, en los experimentos reales los resultados no fueron satisfechos y el cálculo de la política en línea no fue suficientemente rápido. Para poder trabajar en áreas largas, espacio continuo y en línea, proponemos un enfoque, el Continuous Real-time POMCP (CR-POMCP), que hace simulaciones de Monte-Carlo para aprender la política. El método funcionó correctamente en simulación, pero con el robot real resultaba en movimientos lentos en forma zigzag. Por eso, otro método fue propuesto, el cual usa las posiciones con la probabilidad más alta según el mapa de probabilidades (belief). Como la propagación del belief se parece mucho a como funciona un PF, proponemos un método que usa un PF para mantener el belief. El método PF tiene que manejar la falta de observaciones y por eso introducimos una función del peso especial. Los dos métodos para actualizar el belief tienen en cuenta el ruido de los sensores y actuadores, falsos negativos y positivos (durante un periodo corto de tiempo) y obstáculos dinámicos. Finalmente, se presenta un método cooperativo y distribuido para multi agentes, que usa el mapa de probabilidades (belief), y éste usa todas las observaciones. Después, se asigna las posiciones de búsqueda a los agentes, explorando el entorno, y teniendo en cuenta: la probabilidad de la posición, la distancia a la posición y si otro agente ya buscará cerca del lugar. En resumen, la contribución más importante de esta tesis son diferentes métodos de búsqueda y seguimiento de una persona en un entorno urbano con uno o más robots de servicio móviles. Todos estos métodos fueron comprobados en simulación y en experimentación real. ; Zoeken en traceren zijn belangrijke manieren om een mobiele servicerobot te helpen mensen, om te zoeken en te redden en, in het algemeen, om mobiele objecten, dieren of mensen te lokaliseren. En hoewel zoeken evident is voor de mens, is dit niet het geval voor robots. Het vereist namelijk verschillende vaardigeheden zoals verkennen, omgaan met ruis in de sensoren en dynamische obstakels en coördinatie in het geval van meerdere agenten. In deze thesis presenteren we verschillende methoden om een persoon te zoeken en te volgen (search-and-track) in een stedelijke omgeving. De methoden zijn eerst getest met simulaties en daarna met twee echte mobiele servicerobots, genaamd Tibi en Dabo. De robots hebben laser rangefinders om te navigeren en om obstakels en benen van mensen te detecteren. Omdat we ons richten op het zoeken en volgen van mensen gebruiken we bestaande methoden voor de robotnavigatie, detectie van personen en het herkennen van de gezochte persoon. De eerste testen gedaan met het spel hide-and-seek, waar de robot leert om de persoon te vinden. Een Mixed Observable Markov Decision Process (MOMDP) is gebruikt, waar de positie van de robot (zoeker) volledig waarneembaar is en die van de persoon (verstopper) gedeeltelijk. Omdat de berekeningscomplexiteit van het aantal staten afhangt, stellen we een hiërarchische online methode voor die de staatruimte verkleind door het groeperen van staten. Hoewel de methode goed werkt in simulatie, waren de resultaten van de experimenten met de robots minder goed. Verder was het online genereren van de politiek niet snel genoeg. Om in grote omgevingen te kunnen werken en continue staatruimte te gebruiken in real-time, hebben we de Continuous Real-time POMCP (CR-POMCP) voorgesteld. Dit algoritme gebruikt Monte-Carlo simulaties om de politiek (beste actie voor een bepaalde staat) te leren. De methode werkte goed in de simulaties, maar de robot maakte in de experimenten zigzaggende bewegingen. Om deze reden hebben we een nieuwe methode voorgesteld die gebruik maakt van een waarschijnlijkheidskaart (belief) over de lokatie van de persoon. Omdat de propagatie van de POMCP lijkt op hoe een Particle Filter (PF) werkt, stellen we ook een methode voor die een PF gebruikt om de belief bij te houden. De PF methode moet om kunnen gaan met missende observaties en we introduceren daarom een speciale gewichtsfunctie. Beide belief update methoden houden rekening met sensor en actuator ruis, incorrecte negatieve detecties, incorrecte positieve detecties (gedurende een korte tijd) en dynamische obstakels. Als laatst presenteren we een coöperatieve multi-agent methode die gebruik maakt van de genoemde belief update methoden, maar tevens gebruik maakt van de observaties van alle agenten. Vervolgens worden de zoeklocaties toegekend om de hele omgeving te verkennen waarbij rekening gehouden wordt met: de belief, de afstand tot de zoeklocatie en of een andere agent al dichtbij de lokatie gaat zoeken. Samenvattend zijn de belangrijkste bijdragen van deze thesis de verschillende methoden om personen te zoeken en te volgen in een stedelijke omgeving met één of meer mobiele servicerobots. Alle methoden zijn getest met simulaties en met experimenten met echte robots. ; Postprint (published version)

Searching and tracking are important behaviours for a mobile service robot to assist people, to search-and-rescue and, in general, to locate mobile objects, animals or humans. Even though searching might be evident for humans, for robots it is not, since it requires exploring, handling noisy sensors, coping with dynamic obstacles, and coordination in the case of multiple agents. In this thesis, we present several methods to search and track a person in an urban environment. All methods were fist tested extensively in simulation and then in real-life, using one or two mobile service robots, called Tibi and Dabo. The robots have laser range finders, which are used to navigate, to detect obstacles and to detect people's legs. Since we focus on search-and-track methods, we use existing methods for robot navigation, for people detection and person recognition. First tests are done with the hide-and-seek problem, in which the robot learns to catch the hider. Concretely, a Mixed Observable Markov Decision Process (MOMDP) model is used, in which the seeker's location is fully observable and the hider's location partially observable. Since the computational complexity depends on the number of states, we propose a hierarchical on-line method that reduces the state space by grouping them together. Although the method worked properly in simulation, in the real-life experiments the results were not satisfying and the on-line policy calculation was not fast enough to work in real-time. To handle larger environments, work in continuous state space and run in real-time, we propose to use an approach, the Continuous Real-time POMCP (CR-POMCP), that does Monte-Carlo simulations to learn a policy. The method performed correctly in simulation, but on the real robot it resulted in slow zigzag movements. Therefore, a new method is proposed, which uses the highest probable locations, according to its probability map (belief). Since the belief propagation of the POMCP resembles how a Particle Filter (PF) works, we also propose a method that uses a PF to maintain the belief. The PF method has to handle lack of observations, therefore, we introduce a special weight function. Both belief update methods take into account sensor and actuator noise, false negative detections, false positive detections (for a short time) and dynamic obstacles. Finally, a cooperative distributed multi-agent method is presented, it makes use of the previous belief update functions, but it uses all the agents' observations. Next, the search locations are assigned by exploring the working environment, taking into account: the belief, the distance to the search location and if another agent already will search close to it. Summarizing, the main contributions of this thesis are several methods to search and track a person in an urban environment with one or more mobile service robots. All these methods have been shown to work through a set of simulations and real-life experiments. ; La cerca i el seguiment de persones són comportaments importants per un robot mòbil de servei per poder assistir, trobar i ajudar als humans, i en general, per localitzar objectes, animals o vianants. Tot i que la cerca és fàcil per als humans, no ho és per a un robot, ja que requereix exploració, maneig de soroll de sensors, fer front als obstacles dinàmics, i la coordinació en el cas de múltiples agents. En aquesta tesi presentem diferents mètodes per a buscar i seguir a una persona en un entorn urbà. Tots els mètodes han estat provats extensivament en simulació i després en el món real, utilitzant dos robots mòbils de servei, la Tibi i en Dabo. Els robots utilitzen sensors làser per a navegar, detectar obstacles i detectar les cames de les persones. Atès que aquest treball es centra en mètodes de cerca i seguiment, s'han usat els mètodes existents per a la navegació del robot, la detecció i el reconeixement de persones. Primerament, s'han fet proves amb el conegut joc del fet i amagar, on el robot aprèn a trobar l'amagador. S'ha fet servir el model Mixed Observable Markov Decision Process (MOMDP), on la posició del trobador és completament visible i la posició de l'amagador és parcialment visible. Degut a que la complexitat computacional depèn del nombre d'estats, es proposa un mètode jeràrquic en línia que redueix l'espai d'estats, tot agrupant-los. Tot i que el mètode va funcionar correctament en simulació, en els experiments reals els resultats no van ser satisfactoris, i el càlcul de la política no va ser prou ràpid com per treballar en temps real. Per tal de fer front a entorns de més superfície, treballar en l'espai continu i executar en temps real, proposem un nou enfocament, el Continuous Real-time POMCP (CRPOMCP), que fa simulacions de Monte-Carlo per aprendre una política. El mètode va funcionar correctament en l'entorn simulat, però a l'entorn real el robot realitzava lents moviments en zig-zag. Per tant, es proposa un mètode nou, que utilitza els llocs amb més alta probabilitat, d'acord amb el seu mapa de probabilitats (belief). Atès que la propagació de les probabilitats en el POMCP és similar al funcionament d'un filtre de partícules (PF), proposem, a més, un mètode que utilitza un PF per mantenir el belief. El mètode de PF ha de manejar la manca d'observacions. Per tant, introduïm una funció del pes especial. Tots dos mètodes de creences tenen en compte el soroll dels sensors i actuadors, la detecció de falsos negatius i positius (per a un curt període de temps) i els obstacles dinàmics. Finalment, es presenta un mètode multi-agent distribuït cooperatiu, que fa ús de les anteriors funcions d'actualització de la creença (belief), i a més utilitza totes les observacions dels agents. En el proper pas, les ubicacions de cerca s'assignen mitjançant l'exploració de l'entorn de treball, tenint en compte la creença, la distància a la ubicació de cerca i si un altre agent ja buscarà a prop d'ella. En resum, les principals contribucions d'aquesta tesi són diversos mètodes per a la cerca i seguiment d'una persona en un entorn urbà amb un o més robots de serveis mòbils. Tots aquests mètodes han demostrat que funcionen a través d'un conjunt de simulacions i experiments en la entorn real dinàmics. ; La búsqueda y el seguimiento de personas son comportamientos importantes para un robot móvil de servicio para poder asistir, buscar y ayudar a la gente, y en general, para localizar un objeto, animal o humano. Aunque la búsqueda puede parecer muy fácil para los humanos, para los robots no lo es, ya que requiere explorar, manejar ruido de sensores, enfrentarse con obstáculos dinámicos y la coordinación en el caso de haber más agentes. En esta tesis, presentamos diferentes métodos para buscar y seguir a una persona en un entorno urbano. Todos los métodos han sido probados excesivamente en simulación y en experimentos reales, usando uno o dos robots móviles de servicio, Tibi y Dabo. Los robots tienen localizadores láser, los cuales se usan para navegar, detectar obstáculos y detectar piernas. Ya que el principal enfoque en este trabajo son los métodes de buscar-y-seguir, utilizamos métodos existentes para la navigación del robot, la detección de personas y el reconecimiento del humano. Las primeras pruebas se hicieron con el juego de escondite, en el cual el robot aprende a buscar el ocultador. Concretamente, usamos un modelo MOMDP, donde la posición del buscador es completamente observable y la posición del ocultador lo es parcialmente. Como la complejidad computacional depende del número de estados, proponemos un método en línea jerárquica, que reduce el espacio de los estados agrupándolos. Aunque el método funcionó correctamente en simulación, en los experimentos reales los resultados no fueron satisfechos y el cálculo de la política en línea no fue suficientemente rápido. Para poder trabajar en áreas largas, espacio continuo y en línea, proponemos un enfoque, el Continuous Real-time POMCP (CR-POMCP), que hace simulaciones de Monte-Carlo para aprender la política. El método funcionó correctamente en simulación, pero con el robot real resultaba en movimientos lentos en forma zigzag. Por eso, otro método fue propuesto, el cual usa las posiciones con la probabilidad más alta según el mapa de probabilidades (belief). Como la propagación del belief se parece mucho a como funciona un PF, proponemos un método que usa un PF para mantener el belief. El método PF tiene que manejar la falta de observaciones y por eso introducimos una función del peso especial. Los dos métodos para actualizar el belief tienen en cuenta el ruido de los sensores y actuadores, falsos negativos y positivos (durante un periodo corto de tiempo) y obstáculos dinámicos. Finalmente, se presenta un método cooperativo y distribuido para multi agentes, que usa el mapa de probabilidades (belief), y éste usa todas las observaciones. Después, se asigna las posiciones de búsqueda a los agentes, explorando el entorno, y teniendo en cuenta: la probabilidad de la posición, la distancia a la posición y si otro agente ya buscará cerca del lugar. En resumen, la contribución más importante de esta tesis son diferentes métodos de búsqueda y seguimiento de una persona en un entorno urbano con uno o más robots de servicio móviles. Todos estos métodos fueron comprobados en simulación y en experimentación real. ; Zoeken en traceren zijn belangrijke manieren om een mobiele servicerobot te helpen mensen, om te zoeken en te redden en, in het algemeen, om mobiele objecten, dieren of mensen te lokaliseren. En hoewel zoeken evident is voor de mens, is dit niet het geval voor robots. Het vereist namelijk verschillende vaardigeheden zoals verkennen, omgaan met ruis in de sensoren en dynamische obstakels en coördinatie in het geval van meerdere agenten. In deze thesis presenteren we verschillende methoden om een persoon te zoeken en te volgen (search-and-track) in een stedelijke omgeving. De methoden zijn eerst getest met simulaties en daarna met twee echte mobiele servicerobots, genaamd Tibi en Dabo. De robots hebben laser rangefinders om te navigeren en om obstakels en benen van mensen te detecteren. Omdat we ons richten op het zoeken en volgen van mensen gebruiken we bestaande methoden voor de robotnavigatie, detectie van personen en het herkennen van de gezochte persoon. De eerste testen gedaan met het spel hide-and-seek, waar de robot leert om de persoon te vinden. Een Mixed Observable Markov Decision Process (MOMDP) is gebruikt, waar de positie van de robot (zoeker) volledig waarneembaar is en die van de persoon (verstopper) gedeeltelijk. Omdat de berekeningscomplexiteit van het aantal staten afhangt, stellen we een hiërarchische online methode voor die de staatruimte verkleind door het groeperen van staten. Hoewel de methode goed werkt in simulatie, waren de resultaten van de experimenten met de robots minder goed. Verder was het online genereren van de politiek niet snel genoeg. Om in grote omgevingen te kunnen werken en continue staatruimte te gebruiken in real-time, hebben we de Continuous Real-time POMCP (CR-POMCP) voorgesteld. Dit algoritme gebruikt Monte-Carlo simulaties om de politiek (beste actie voor een bepaalde staat) te leren. De methode werkte goed in de simulaties, maar de robot maakte in de experimenten zigzaggende bewegingen. Om deze reden hebben we een nieuwe methode voorgesteld die gebruik maakt van een waarschijnlijkheidskaart (belief) over de lokatie van de persoon. Omdat de propagatie van de POMCP lijkt op hoe een Particle Filter (PF) werkt, stellen we ook een methode voor die een PF gebruikt om de belief bij te houden. De PF methode moet om kunnen gaan met missende observaties en we introduceren daarom een speciale gewichtsfunctie. Beide belief update methoden houden rekening met sensor en actuator ruis, incorrecte negatieve detecties, incorrecte positieve detecties (gedurende een korte tijd) en dynamische obstakels. Als laatst presenteren we een coöperatieve multi-agent methode die gebruik maakt van de genoemde belief update methoden, maar tevens gebruik maakt van de observaties van alle agenten. Vervolgens worden de zoeklocaties toegekend om de hele omgeving te verkennen waarbij rekening gehouden wordt met: de belief, de afstand tot de zoeklocatie en of een andere agent al dichtbij de lokatie gaat zoeken. Samenvattend zijn de belangrijkste bijdragen van deze thesis de verschillende methoden om personen te zoeken en te volgen in een stedelijke omgeving met één of meer mobiele servicerobots. Alle methoden zijn getest met simulaties en met experimenten met echte robots. ; Postprint (published version)

[Abstract] This paper addresses the main results obtained during the design and analysis of a cable-driven robot able to simulate the dynamic conditions existing in underwater environment. This work includes the kinematic and dynamic modeling as well as the analysis of the tension of the cables along different trajectories. The low-gravity simulator application is novel in the context of cable-driven robots and it is aimed to be implemented in an underwater humanoid robot. Therefore, this work can be seen as a test case of the complementary research contributions of the group of Robotics and Intelligent Machines at CAR in the recent years. ; The research leading to these results has received funding from the Spanish Government CICYT project Ref. DPI2014-57220-C2-1-P, DPI2013-49527-EXP, the Universidad Politécnica de Madrid project Ref. AL14-PID-15, and the RoboCity2030-III-CM project (Robótica aplicada a la mejora de la calidad de vida de los ciudadanos. Fase III; S2013/MIT-2748), funded by Programas de Actividades I+D en la Comunidad de Madrid and cofunded by Structural Funds of the EU ; Universidad Politécnica de Madrid; AL14-PID-15 ; Comunidad de Madrid; S2013/MIT-2748 ; https://doi.org/10.17979/spudc.9788497498081

Generating complex, human-like behavior in a humanoid robot like the iCub requires the integration of a wide range of open source components and a scalable cognitive architecture. Hence, we present the iCub-HRI library which provides convenience wrappers for components related to perception (object recognition, agent tracking, speech recognition, and touch detection), object manipulation (basic and complex motor actions), and social interaction (speech synthesis and joint attention) exposed as a C++ library with bindings for Java (allowing to use iCub-HRI within Matlab) and Python. In addition to previously integrated components, the library allows for simple extension to new components and rapid prototyping by adapting to changes in interfaces between components. We also provide a set of modules which make use of the library, such as a high-level knowledge acquisition module and an action recognition module. The proposed architecture has been successfully employed for a complex human–robot interaction scenario involving the acquisition of language capabilities, execution of goal-oriented behavior and expression of a verbal narrative of the robot's experience in the world. Accompanying this paper is a tutorial which allows a subset of this interaction to be reproduced. The architecture is aimed at researchers familiarizing themselves with the iCub ecosystem, as well as expert users, and we expect the library to be widely used in the iCub community. ; The research leading to these results has received funding under the European Union's Seventh Framework Programme (FP/2007-2013)/ERC Grant Agreement n. FP7-ICT-612139 (WYSIWYD—What You Say Is What You Did) and FP7-ICT-270490 (EFAA—The Experimental Functional Android Assistant). PN was supported by a Marie Curie Early Stage Researcher Fellowship (H2020-MSCA-ITA, SECURE 642667). PV was supported by the ERC advanced grant 341196 (cDAC—Role of Consciousness in Adaptive Behavior).

In this paper, we present a novel methodology to obtain imitative and innovative postural movements in a humanoid based on human demonstrations in a different kinematic scale. We collected motion data from a group of human participants standing up from a chair. Modeling the human as an actuated 3-link kinematic chain, and by defining a multi-objective reward function of zero moment point and joint torques to represent the stability and effort, we computed reward profiles for each demonstration. Since individual reward profiles show variability across demonstrating trials, the underlying state transition probabilities were modeled using a Markov chain. Based on the argument that the reward profiles of the robot should show the same temporal structure of those of the human, we used differential evolution to compute a trajectory that fits all humanoid constraints and minimizes the difference between the robot reward profile and the predicted profile if the robot imitates the human. Therefore, robotic imitation involves developing a policy that results in a temporal reward structure, matching that of a group of human demonstrators across an array of demonstrations. Skill innovation was achieved by optimizing a signed reward error after imitation was achieved. Experimental results using the humanoid HOAP-3 are shown. ; The research leading to these results has received funding from the ARCADIA project DPI2010-21047-C02-01, funded by CICYT project grant on behalf of Spanish Ministry of Economy and Competitiveness and from the RoboCity2030-II-CM project (S2009/DPI-1559), funded by Programas de Actividades I+D en la Comunidad de Madrid and co-funded by Structural Funds of the EU. ; Publicado

This is the author accepted manuscript. The final version is available from Springer Nature via the DOI in this record. ; Rapid progress in the area of humanoid robots offers tremendous possibilities for investigating and improving social competences in people with social deficits, but remains yet unexplored in schizophrenia. In this study, we examined the influence of social feedbacks elicited by a humanoid robot on motor coordination during a human-robot interaction. Twenty-two schizophrenia patients and twenty-two matched healthy controls underwent a collaborative motor synchrony task with the iCub humanoid robot. Results revealed that positive social feedback had a facilitatory effect on motor coordination in the control participants compared to non-social positive feedback. This facilitatory effect was not present in schizophrenia patients, whose social-motor coordination was similarly impaired in social and non-social feedback conditions. Furthermore, patients' cognitive flexibility impairment and antipsychotic dosing were negatively correlated with patients' ability to synchronize hand movements with iCub. Overall, our findings reveal that patients have marked difficulties to exploit facial social cues elicited by a humanoid robot to modulate their motor coordination during human-robot interaction, partly accounted for by cognitive deficits and medication. This study opens new perspectives for comprehension of social deficits in this mental disorder. ; This experiment was supported by a grant from the European Union's Seventh Framework Program (FP7 ICT 2011 Call 9) under grant agreement n FP7-ICT-600610 ALTEREGO.

Technological advances in the field of robotics have entered various aspects of human life ranging from automation in the fields of industry, military, entertainment, and in the medical field. Humanoid robot or humanoid robot is a form of technology implementation in the field of robotics that can mimic some human activities to walk and move several body organs like humans. The design of Jaipong Dance Robot based on CM-530 consists of a minimum system that functions as a controller of servo motor movement as a robot actuator and processes the data generated by the gyro sensor as a balance sensor. The design of Jaipong Dance Robot based on CM-530 has been successfully made and can work well, from standing movements, opening 1, opening 2, pencugan 1, pencugan 2, mincid kendor 1, pencugan 3, mincid kendor 2, mincid gancang 1, mincid gancang 2, mincid gancang 3, mincid gancang 4, nibakeun, dan mincid gancang 5, the success rate is 80%. Gyroscope GS-12 sensor can provide the desired balance point, the Dynamixel AX-12A Servo Motor as propulsion which is located at each joint of the jaipong dance robot, the Battery / Power Supply as the power provider. ; Kemajuan teknologi dalam bidang robotika telah memasuki berbagai segi kehidupan manusia mulai dari bidang otomatisasi baik di bidang industri, militer, hiburan maupun dalam bidang medis. Humanoid robot atau robot humanoid merupakan salah satu bentuk implementasi teknologi dalam bidang robotika yang memiliki kemampuan menirukan beberapa kegiatan manusia untuk berjalan dan menggerakan beberapa organ tubuh layaknya manusia. Perancangan Robot Tari Jaipong berbasis CM-530 terdiri dari sistem minimum yang berfungsi sebagai pengontrol pergerakan motor servo sebagai aktuator robot dan mengolah data yang dihasilkan sensor gyro sebagai sensor keseimbangan. Perancangan Robot Tari Jaipong berbasis CM-530 sudah berhasil dibuat dan dapat bekerja dengan baik, dari gerakan berdiri, bukaan 1, bukaan 2, pencugan 1, pencugan 2, mincid kendor 1, pencugan 3, mincid kendor 2, mincid gancang 1, mincid gancang 2, mincid gancang 3, mincid gancang 4, nibakeun, dan mincid gancang 5 tingkat keberhasilan adalah 80%. Sensor Gyroscope GS-12 dapat memberikan titik keseimbangan yang diinginkan, Motor Servo Dynamixel AX-12A sebagai alat penggerak yang mana terletak pada setiap sendi robot tari jaipong, Baterai / Power Supply sebagai pemberi daya.

Trabajo presentado en el 19th International Conference on Humanoid Robots (Humanoids), celebrado en Toronto (Canadá) del 15 al 17 de octubre de 2019 ; Current locomotion algorithms in structured (in-door) 3D environments require an accurate localization. The several and diverse sensors typically embedded on legged robots (IMU, coders, vision and/or LIDARS) should make it possible if properly fused. Yet this is a difficult task due to the hetero-geneity of these sensors and the real-time requirement of the control. While previous works were using staggered approaches (odometry at high frequency, sparsely corrected from vision and LIDAR localization), the recent progress in optimal estimation, in particular in visual-inertial localization, is paving the way to a holistic fusion. This paper is a contribution in this direction. We propose to quantify how a visual-inertial navigation system can accurately localize a humanoid robot in a 3D indoor environment tagged with fiducial markers. We introduce a theoretical contribution strengthening the formulation of Forster's IMU pre-integration, a practical contribution to avoid possible ambiguity raised by pose estimation of fiducial markers, and an experimental contribution on a humanoid dataset with ground truth. Our system is able to localize the robot with less than 2 cm errors once the environment is properly mapped. This would naturally extend to additional measurements corresponding to leg odometry (kinematic factors) thanks to the genericity of the proposed pre-integration algebra. ; This work has been supported by the MEMMO European Union project within the H2020 Program under Grant Agreement No. 780684, the Spanish Ministry of Science, Innovation, and Universities project EB-SLAM (DPI2017-89564-P), by the EU H2020 project GAUSS (DPI2017-89564-P) and by the Spanish State Research Agency through the Maria de Maeztu Seal of Excellence to IRI MDM-2016-0656.

A 500Kg, self-contained biped robot, named Roboshift, has been conceived and tested to investigate issues associated with the control of industrial scale biped robots. This project represents the first credible attempt to build a heavy weight autonomous biped robot. The recent expansion in humanoid robot development has highlighted advances made in anthropomorphic biped technology. Current research into speech recognition, vision systems, laser topography, artificial intelligence and electroactive polymers will ultimately achieve an Android capable of human like actions and thought processes. Justification for this most demanding and expensive research is based on philanthropic models that suggest these robots will attend to the bedridden, or replace humans in dangerous areas. However, the cost of a biped robot when compared to that of a wheeled or tracked vehicle restricts commercialisation for these applications. As well, the size and working capacity of current humanoid robots is not compatible with the heavy lifting requirements found in such environments. It is proposed that only biped robots of an industrial scale, possessing a capacity much greater than that of a human, will be of commercial value in the future. Typical applications may include the handling of materials in confined or uneven terrain, where a forklift or other commercially available materials handling equipment would be unsuitable. For example, field handling in military, mining or geological environments. Minimal research has been conducted into the realisation of such a device, which presents challenges in terms of the magnitude of dynamic forces produced and of the systems required to control the robot in real-time. Review of relevant literature reveals that little research has been completed in this field. Therefore, operational characteristics for an industrial scale biped robot are defined. The design then details the structure and integration of mechanical, hydraulic, and electrical systems. Roboshift is powered by an internal ...

In the field of robotics, there is a growing awareness of the importance of benchmarking [1], [2]. Benchmarking not only allows the assessment and comparison of the performance of different technologies but also defines and supports the standardization and regulation processes during their introduction to the market. Its importance has been recently emphasized by the adoption of the technology readiness levels (TRLs) in the Horizon 2020 information and communication technologies by the European Union as an important guideline to assess when a technology can shift from one TRL to the other. The objective of this article is to define the basis of a benchmarking scheme for the assessment of bipedal locomotion that could be applied and shared across different research communities. ; European Commission Seventh Framework Program, and COST

This study suggests an innovative way of using an intelligent robot to support speech therapy for hearing impaired children through play. Although medical technology (e.g., hearing aid, cochlear implant) for children with hear-ing impairment has advanced significantly, the amplification itself does not provide optimal development of hearing and speaking; it must be combined with specialized therapy. The present study focuses on the use of the humanoid robot NAO in auditory-verbal therapy, an approach to the development of audi-tory and verbal skills, which does not allow lipreading or other non-verbal cues to facilitate communication. NAO does not have a human mouth and therefore children with hearing impairment cannot do lipreading; this unique characteris-tic of the technology has been successfully used in the study to create playful and engaging auditory-verbal therapy sessions for six kindergarten hearing im-paired children, allowing them to improve their ability to follow instructions us-ing the hearing aid/ cochlear implant rather than visual clues in the environ-ment. Our results, although preliminary, seem to encourage further research in supporting hearing-impaired children via play with intelligent robots. ; This work has been partly supported by the project that has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 739578 (RISE – Call: H2020-WIDESPREAD-01-2016-2017-TeamingPhase2) and the Government of the Republic of Cyprus through the Directorate General for European Programmes, Coordination and Development.

The version of record is available online at: https://doi.org/10.1007/s10514-021-09976-7 ; This paper proposes to enrich robot motion data with trajectory curvature information. To do so,we use an approximate implementation of a topological feature named writhe, which measures the curling of a closed curve around itself, and its analog feature for two closed curves, namely the linking number. Despite these features have been established for closed curves, their definition allows for a discrete calculation that is well-defined for non-closed curves and can thus provide information about how much a robot trajectory is curling around a line in space. Such lines can be predefined by a user, observed by vision or, in our case, inferred as virtual lines in space around which the robot motion is curling. We use these topological features to augment the data of a trajectory encapsulated as a Movement Primitive (MP). We propose a method to determine how many virtual segments best characterize a trajectory and then find such segments. This results in a generative model that permits modulating curvature to generate new samples, while still staying within the dataset distribution and being able to adapt to contextual variables. ; This work has been carried out within the project CLOTHILDE ("CLOTH manIpulation Learning from DEmonstrations") funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Advanced Grant agreement No 741930). Research at IRI is also supported by the Spanish State Research Agency through the Mar ́ıa de Maeztu Seal of Excellence to IRI MDM-2016-0656 ; Peer Reviewed ; Postprint (author's final draft)

© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. ; Learning from Demonstration (LfD) is a paradigm that allows robots to learn complex manipulation tasks that can not be easily scripted, but can be demonstrated by a human teacher. One of the challenges of LfD is to enable robots to acquire skills that can be adapted to different scenarios. In this paper, we propose to achieve this by exploiting the variations in the demonstrations to retrieve an adaptive and robust policy, using Gaussian Process (GP) models. Adaptability is enhanced by incorporating task parameters into the model, which encode different specifications within the same task. With our formulation, these parameters can be either real, integer, or categorical. Furthermore, we propose a GP design that exploits the structure of replications, i.e., repeated demonstrations with identical conditions within data. Our method significantly reduces the computational cost of model fitting in complex tasks, where replications are essential to obtain a robust model. We illustrate our approach through several experiments on a handwritten letter demonstration dataset. ; This work has been partially funded by the European Union Horizon 2020 Programme under grant agreement no. 741930 (CLOTHILDE) and by the Spanish State Research Agency through the Mar ́ıa de Maeztu Seal of Excellence to IRI [MDM-2016-0656]. ; Peer Reviewed ; Postprint (author's final draft)

According to the Swedish science fiction TV series Äkta människor (Real Humans, SVT and Matador film 2012-2014), humanoid robots called "hubots" are replacing the human workforce in care work and assembly line industries. Against the backdrop of current debates about immigration and citizenship in the Nordic countries, this article does a close, contextual reading of the series, exploring how the hubots influence work and family life. We are particularly interested in how hubots tie in with the cultural circulation of aff ect in relation to Otherness and how responses towards the "not-quite" human or dehumanized Other are negotiated in present-day Nordic cultural imaginaries. What kinds of affects are at stake in how Äkta människor takes up and interacts with debates about immigrant workers and the "not-quite" human? To answer these questions, the article develops the notion of "affective imaginaries" as an analytical tool for understanding the exchange between popular culture and political debate. ; publishedVersion