Search

Underwater 3-D laser scanners are an essential type of sensors used by unmanned underwater vehicle (UUVs) for operations such as inspection, navigation, and object recognition and manipulation. This article presents a novel 3-D laser scanner, which uses a 2-axis mirror to project straight lines into the water by compensating for refraction-related distortions. This is achieved by projecting optimally curved lines, so that the refraction when they enter the water transforms them into straight lines. The relevance of this approach lies in the fact that 3-D triangulation using planes is noticeably faster than using elliptic cones. The goal of this work is twofold: first, to prove that refraction-related distortions can in practice be compensated for by using a 2-axis mirror, and second, to present a simple calibration algorithm that only needs to compute the coefficients of polynomial functions. To the best of the authors’ knowledge, the prototype presented in this article is the first laser line scanner that actively counteracts the refraction of the projected light in the context of underwater robotics

Aquest treball aborda el disseny i muntatge d’un prototip i la creació d’un sistema original capaç de mesurar la distància a una font sonora exclusivament a partir del senyal captat per un o més micròfons i tractant aquest senyal de manera adequada, amb l’objectiu de poder aplicar aquest sistema a la determinació de l’angle existent entre un robot mòbil i una persona que interaccioni amb ell amb unes distàncies predeterminades. El senyal de la font és d’una sola freqüència, també coneguda. El disseny del prototip consta d’una primera etapa analògica, que du a terme l’amplificació, el filtrat i la rectificació del senyal, i d’una segona etapa digital que s’encarrega de convertir la lectura del nivell de pressió sonora captat a voltatge mesurable. Paral·lelament al desenvolupament del prototip hi ha una important part de presa de mesures i d’estudi de les dades obtingudes d’un model teòric com pràctic per tal d’estudiar l’abast de la fórmula de conversió que s’utilitzarà. Finalment es comprovarà el funcionament del prototip dissenyat. Com a requisit per a poder desenvolupar el projecte es contempla la recerca d’informació i l’estudi de les diverses tecnologies i aparells, així com del marc teòric que l’engloba, és a dir, tot el referent a com es transmet el so i les diverses maneres de caracteritzar-lo Este trabajo aborda el diseño y montaje de un prototipo y la creación de un sistema original capaz de mesurar la distancia a una fuente sonora exclusivamente a partir de la señal captada por uno o más micrófonos y tratando esta señal de manera adecuada, con el objetivo de poder aplicar este sistema a la determinación del ángulo existente entre un robot móvil y una persona que interaccione con él con unas distancias predeterminadas. La señal de la fuente es de una sola frecuencia, también conocida. El diseño del prototipo consta de una primera etapa analógica, que lleva a cabo la amplificación, el filtrado y la rectificación de la señal, y de una segunda etapa digital que se encarga de convertir la lectura del nivel de presión sonora captado a voltaje medible. Paralelamente al desarrollo del prototipo, hay una importante parte de toma de medidas y de estudio de los datos obtenidos de un modelo tanto teórico como práctico para estudiar el alcance de la fórmula de conversión que se utilizará. Finalmente se comprobará el funcionamiento del prototipo diseñado. Como requisito para poder desarrollar el proyecto se contempla la investigación de información y el estudio de las varias tecnologías y aparatos, así como del marco teórico que lo engloba, es decir, todo lo en lo referente a cómo se transmite el sonido y las diversas maneras de caracterizarlo This work deals with the design and assembly of a prototype and the creation of an original system capable of measuring the distance to a sound source exclusively from the signal captured by one or more microphones and treating this signal adequately, with the aim of applying this system to the determination of the existing angle between a moving robot and a person interacting with it with predetermined distances. The source signal is a single frequency, also known. The prototype design consists of a first stage analogue, which performs amplified, filtered and corrected signal, and a second stage digital which converts the reading of the level of sound pressure captured to measurable voltage. In parallel to the development of the prototype, there is an important part of taking measures and studying the data obtained from a theoretical as practical model in order to study the scope of the conversion formula to be used. Finally, the operation of the designed prototype will be checked. Research into information and the study of the various technologies and devices and the theoretical framework that includes it are a prerequisite for the development of the project, everything relating to how sound is transmitted and the various ways of characterizing it

El Robotino es un robot mòbil desenvolupat per Festo Didactics per a l’aprenentatge i la investigació en el camp dels robots autònoms. El problema del posicionament d’aquest tipus de robots mòbils és àmpliament estudiat i encara no té una solució ideal per a totes les àrees d’aplicació. El Robotino utilitza la tècnica de l’odometria amb les seves rodes per estimar la posició i orientació. Per millorar la precisió del posicionament, anteriorment s’han desenvolupat altres projectes, com la integració d’un sensor giroscopi per estimar la seva orientació amb més precisió. L’objectiu d’aquest projecte es millorar les capacitats de posicionament del Robotino. Per a fer-ho, s’ha desenvolupat una solució de posicionament alternativa basada en sistemes de navegació inercial (giroscopis i acceleròmetres). El dispositiu escollit és una Unitat de Mesura Inercial (IMU), que s’ha integrat físicament amb el Robotino a traves del port sèrie RS-232. S’ha desenvolupat el programari per fer servir aquest nou dispositiu juntament amb el programari existent del Robotino, el qual calcula la velocitat i posició del robot en tot moment. S’han realitzat proves per verificar l’eficàcia del nou sistema de posicionament, les quals han consistit a fer que el Robotino realitzi diferents desplaçaments i extreure’n les dades cinemàtiques per comparar-les amb el sistema de posicionament per odometria. Les proves s’han fet amb un altre programari d’aplicació demostrativa, desenvolupada en aquest projecte. Els resultats obtinguts mostren que el nou sistema de posicionament no es millor que el sistema anterior, i per tant no s’ha assolit l’objectiu del treball. Els motius d’això han estat la falta de precisió en la calibració del dispositiu utilitzat, així com les limitacions pròpies del dispositiu per a aquest tipus d’aplicacions

Per a moltes empreses, l'automatització dels magatzems s'està convertint en un procés clau per millorar l'eficiència, reduir els costos i augmentar la seguretat. En aquest procés, els robots AGV son una eina fonamental. L'objectiu d'aquest projecte era desenvolupar un sistema de gestió i control remot per a un robot AGV (vehicle guiat autònomament) per a la seva utilització futura en aplicacions d'automatització de magatzem. Durant la realització del projecte, s'ha complert aquest objectiu amb èxit gràcies a la utilització de dades en temps real de lidar, càmera i altres sensors de l'AGV. Això ha permès al robot navegar pel magatzem, transportar mercaderia i inventariar el magatzem automàticament. For many companies, warehouse automation is becoming a key process to improve efficiency, reduce costs and increase security. As small, autonomous robotic platforms, AGV robots are key enablers for warehouse automation processes. The objective of this project was to develop a remote control and management system for an AGV robot (autonomous guided vehicle) for its future use in warehouse automation applications. During the realization of the project, this objective has been successfully met thanks to the use of real-time data from Lidar, camera and other AGV sensors that, with the developed system, allow the AGV robot to navigate through the warehouse, transport merchandise and automatically log or update warehouse inventory Para muchas empresas, la automatización de los almacenes se está convirtiendo en un proceso clave para mejorar la eficiencia, reducir los costes y aumentar la seguridad. En estos procesos de automatización, los robots AGV son una herramienta fundamental. El objetivo de este proyecto es desarrollar un sistema de gestión y control remoto para un robot AGV (vehículo guiado autónomamente) para su futura utilización en aplicaciones de automatización de almacén. Durante la realización del proyecto se ha cumplido este objetivo con éxito gracias a la utilización de datos en tiempo real de Lidar, cámara y otros sensores del AGV que con el sistema desarrollado permiten al robot AGV navegar por el almacén, transportar mercancía e inventariar el almacén automáticamente

Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2023, Director: Jordi Vitrià i Marca, Sergi Hernández i Alejandro López, [en] This project aims to supply the ADD autonomous robot with a landmark detection system that should be able to easily integrate with the existing localization stack. Multiple landmark have been developed, but this project is centered on urban traffic signs. To be able to detect and classify the signs, three different methods are proposed: the first uses a convolutional neural network (CNN) to both detect and classify the signs; the second one uses a CNN to detect and another one to classify; the third, uses a CNN to detect and more traditional methods to classify. Multiple networks and optimizations are considered, with a strong emphasis with those on the YOLO family. These methods have been tested with the well known GTSDB and GTSRB, and some results have been comparable to the ones obtained in the state of the art. They are also tested in environments much more similar to the ones the robot will encounter, using a specially made dataset.

One of the main goals in robotics is to achieve full autonomy. However, for certain tasks, robots still lack the level of abstract reasoning that gives the ability to make decisions, plan ahead, and change actions during the execution of a process, these abilities are acquired at its best degree only by humans. In terms of unknown underwater environments, the combination of specialized robotics that can withstand harsh environments and the reasoning of a human operator can give effective results on exploration, inspection, and subsequent tasks. Standard solutions propose to use fully teleoperated Remotely Operated Vehicles (ROV)s or fully Autonomous Underwater Vehicles (AUV). Still, both solutions have their drawbacks: high operation costs and limitations due to the need of a physical connection in the case of ROVs, and limitations in the communications or problems while facing complex decision making in case of AUVs. These limitations can be overcome with the concept of shared autonomy applied to a Hybrid ROV (HROV). In this new paradigm, an operator leads the mission by selecting objectives at a high level, and a HROV executes them taking all the low level decisions. This paper explores the shared autonomy concept applied to an underwater exploration task, leading the first steps towards intervention missions.

Aquest projecte centra la seva atenció en el desenvolupament d’una aplicació per a poder entrenar un robot per a que pugui dur a terme una tasca de Pick and Place. En el món de l’aprenentatge automàtic, existeixen diferents algorismes d’implementació per a poder fer exercicis d’aquest tipus, en aquest treball s’ha triat el Deep Deterministic Policy Gradient (DDPG) fusionat amb el Hindsight Experience Replay (HER). S’ha utilitzat el gym d’OpenAI per a provar i executar els diferents fulls de codi que s’han desenvolupat amb el llenguatge de programació de Python. Al llarg de l’estudi s’exposen les bases teòriques a seguir i s’expliquen les diferents estructures dels scripts desenvolupats per a entrenar al model programat. D’aquesta manera es pot seguir el fil del disseny la IA de principi a final adquirint també, paral·lelament, coneixement del món de la intel·ligència artificial. El proyecto presente centra su atención en el desarrollo de una aplicación para poder entrenar un robot para que pueda realizar una tarea de Pick and Place. En el mundo del aprendizaje autónomo, existen diferentes algoritmos de implementación para poder hacer ejercicios de este tipo, en este trabajo se ha escogido a Deep Deterministic Policy Gradient (DDPG) mezclado con el Hindsight Experience Replay (HER). Se ha utilizado el gym de OpenAI para probar y ejecutar las distintas hojas de código que se han desarrollado con el lenguaje de programación de Python. A lo largo del estudio se explican las bases teóricas a seguir y se exponen las diferentes estructures de los scripts creados para entrenar al modelo programado. De esta forma, se puede seguir el hilo del diseño de la IA de principio a fin, adquiriendo también, paralelamente, conocimientos del mundo de la inteligencia artificial. This project focuses on the development of an application to train a robot to perform a Pick and Place task. In the world of autonomous learning, there are different implementation algorithms to be able to do exercises of this type, in this work Deep Deterministic Policy Gradient (DDPG) has been chosen and mixed with Hindsight Experience Replay (HER). The OpenAI gym has been used to test and execute the different code sheets that have been developed with the Python programming language. Throughout the study, the theoretical bases to be followed are explained and the different structures of the scripts created to train the programmed model are presented. In this way, it is possible to follow the AI design thread from start to finish, acquiring, in parallel, knowledge of the world of artificial intelligence.

L’ús de sistemes autònoms amb bateries recarregables és cada cop més estès a la indústria, especialment en aplicacions robòtiques. Les noves plataformes robòtiques requereixen una alta autonomia per poder allargar la durada dels processos, a més de la capacitat de predir quan s’esgotarà l’energia per a carregar el sistema, avortar tasques, o tornar a una estació base. Per aquest motiu, la correcta estimació del State-of-Charge (SoC) per poder monitorar l’estat de la bateria i garantir un correcte funcionament del dispositiu és una aplicació altament demandada. En aquest treball es proposa desenvolupar, implementar i testejar un sistema que estimi el SoC d’una bateria i enviar-lo a un ordinador central perquè, en cada cas, prengui la decisió oportuna. El resultat final es provarà en una plataforma autònoma real de l’empresa PAL Robotics. El projecte desenvoluparà la programació de l’algoritme per monitorar el nivell de càrrega de les bateries. Aquestes dades seran transmeses a l’ordinador central mitjançant el protocol Universal Serial Bus (USB), juntament amb altres dades del robot. D’aquesta manera, en aquest document s’explica detalladament com s’han desenvolupat les tasques que permeten caracteritzar i validar els drivers per PAL Robotics. Als annexes s’inclouen els header files de la solució proposada així com també els scripts i experiments que han permès desenvolupar i/o comprovar el funcionament d’aquests. Després d’haver realitzat aquestes tasques, s’ha pogut comprovar que la feina implementada en aquest TFG és vàlida donat que es superen els requeriments tècnics demanats per l’empresa. El uso de sistemas autónomos con baterías recargables es cada vez más extendido en la industria, especialmente en aplicaciones robóticas. Las nuevas plataformas robóticas requieren una alta autonomía para poder alargar la duración de los procesos, además de la capacidad de predecir cuándo se agotará la energía para cargar el sistema, abortar tareas, o volver a una estación base. Por este motivo, la correcta estimación del State-of-Charge (SoC) para poder monitorizar el estado de la batería y garantizar un correcto funcionamiento del dispositivo es una aplicación altamente demandada. En este trabajo se propone desarrollar, implementar y testear un sistema que estime el SoC de una batería y enviarlo a un ordenador central para que, en cada caso, tome la decisión oportuna. El resultado final se probará en una plataforma autónoma real de la empresa PAL Robotics. El proyecto desarrollará la programación del algoritmo para monitorizar el nivel de carga de las baterías. Estos datos serán transmitidos al ordenador central mediante el protocolo Universal Serial Bus (USB), junto con otros datos del robot. De este modo, en este documento se explica detalladamente cómo se han desarrollado las tareas que permiten caracterizar y validar los drivers por PAL Robotics. En los anexos se incluyen los header files de la solución propuesta, así como los scripts y experimentos que han permitido desarrollar y/o comprobar el funcionamiento de los mismos. Después de haber realizado estas tareas, se ha podido comprobar que el trabajo implementado en este TFG es válido dado que se superan los requerimientos técnicos solicitados por la empresa. The use of autonomous systems with rechargeable batteries is becoming more widespread in the industry, especially in robotic applications. New robotic platforms require high autonomy to be able to extend the duration of processes, as well as the ability to predict when the power will run out to charge the system, abort tasks, or return to a base station. For this reason, the correct estimation of the State-of-Charge (SoC) to be capable to monitor the battery status and guarantee the correct operation of the device is a highly demanded application. In this work, we propose to develop, implement and test a system that estimates the SoC of a battery and sends it to a central computer to make the appropriate decision in each case. The proposal will be tested on a real autonomous platform of the company PAL Robotics. The project will develop the programming of the algorithm to monitor the charge level of the batteries. This data will be transmitted to the host computer via Universal Serial Bus (USB) protocol and other robot data. In this way, this document explains in detail how the tasks that allow the drivers to be characterized and validated by PAL Robotics have been developed. The annexes include the header files of the proposed solution, as well as the scripts and experiments that have allowed to develop and/or test the operation of the same. After having carried out all these experiments, it has been possible to verify that the work implemented in this TFG is valid since the technical requirements requested by the company are exceeded.

Aquest document presenta un Treball de Fi de Grau on es realitza el disseny, construcció i programació d’un vehicle autònom, és a dir, un robot mòbil capaç de desplaçar-se d’una posició a una altra sense la necessitat d’haver de ser controlat de manera manual. El vehicle esmentat és impulsat per la Jetson Nano, un ordinador d’una sola placa (Single-Board Computer o SBC) ideal per aplicacions de visió per computador, intel·ligència artificial, etc., i que permet al vehicle tenir autonomia gràcies al tractament d’imatges. Addicionalment, el robot pot ser guiat manualment a través de diferents dispositius com ordinadors o telèfons mòbils, mentre s’observa la imatge en temps real que aquest proporciona i altres dades de sensors en una interfície gràfica d’usuari. Per tal que el vehicle sigui funcional, es realitza el disseny mecànic de la seva estructura, el disseny electrònic dels components que fan possible el seu moviment i percepció de l’entorn, i el disseny del software que li aporta la intel·ligència necessària per ser autònom. Posteriorment, es realitza la seva construcció i la programació d’una aplicació per comprovar el seu correcte funcionament. Finalment, s’observa que el vehicle pot tenir múltiples aplicacions en l’àmbit de l’educació i la investigació i que és una millor proposta que altres possibilitats que hi ha al mercat. Este documento presenta un Trabajo de Fin de Grado donde se realiza el diseño, la construcción y programación de un vehículo autónomo, es decir, un robot móvil capaz de desplazarse de una posición a otra sin la necesidad de tener que ser controlado de manera manual. El vehículo mencionado es impulsado por la Jetson Nano, un ordenador de una sola placa (Single-Board Computer o SBC) ideal para aplicaciones de visión por computador, inteligencia artificial, etc., y que permite al vehículo tener autonomía gracias al tratamiento de imágenes. Adicionalmente, el robot puede ser guiado manualmente a través de diferentes dispositivos como ordenadores o teléfonos móviles, mientras se observa la imagen en tiempo real que éste proporciona y otros datos de sensores en una interfaz gráfica de usuario. A fin de que el vehículo sea funcional, se realiza el diseño mecánico de su estructura, el diseño electrónico de los componentes que hacen posible su movimiento y percepción del entorno, y el diseño del software que le aporta la inteligencia necesaria para ser autónomo. Posteriormente, se realiza su construcción y la programación de una aplicación para comprobar su correcto funcionamiento. Finalmente, se observa que el vehículo puede tener múltiples aplicaciones en el ámbito de la educación y la investigación y que es una mejor propuesta que otras posibilidades que hay en el mercado. This document presents a Final Bachelor’s Degree Project where the design, building and programming of an autonomous vehicle are carried out, that is, a mobile robot capable of moving from one position to another without the need of being controlled in manual mode. The aforementioned vehicle is powered by the Jetson Nano, an ideal Single-Board Computer (SBC) for applications of computer vision, artificial intelligence, etc., and which allows the vehicle to have autonomy thanks to the image processing. Additionally, the robot may be guided manually from different devices such as computers or mobile phones, while the real time image that it provides and other sensors data are observed in a graphical user interface. In order for the vehicle to be functional, the mechanical design of its structure, the electronic design of the components that make possible its movement and environment perception, and the software design that provides its intelligence to be autonomous are carried out. Later, the robot is built and the programming of an application in order to test its correct behaviour is done. Finally, it has been observed that the vehicle can have multiple applications in the education and investigation areas and that it is a better proposal than other possibilities able in the market.

Underwater 3D laser scanners are an essential type of sensor used by unmanned underwater vehicles (UUVs) for operations such as navigation, inspection and object recognition and manipulation. These sensors need to be able to provide highly accurate 3D data at fast refresh rates in order to accomplish these tasks. Usually, these scanners rely on a rotating mirror actuated by a galvanometer. However, the light planes steered by this type of mirrors are typically deformed into cones due to refraction. In order to produce accurate results, this distortion needs to be taken into account, which increases the computational cost of the 3D reconstruction. A novel approach consisting in using a biaxial MEMS mirror is proposed in this paper. The second rotational degree of freedom of the mirror can be used to project optimally curved light shapes, so that the refraction process transforms them into planes. Being able to model the light surfaces as planes rather than cones can significantly reduce the computation time of the 3D reconstruction. In order to do so, an exhaustive model of the complete light trajectories is presented. To the best of the authors' knowledge, this paper constitutes the first attempt to model and counteract the distortion in the scanning pattern introduced by a biaxial mirror and a double refraction process in the context of underwater robotics This work was supported in part by the Spanish Ministry of Science through the GIRONA1000 Project under Grant DPI2017-86372-C3-2-R, in part by the European Commission through the ATLANTIS Project under Grant H2020-ICT-2019-2-871571, and in part by the Doctoral Grant of the University of Girona under Grant IFUdG2019

Underwater Inspection, Maintenance, and Repair operations are nowadays performed using Remotely Operated Vehicles (ROV) deployed from dynamic-positioning vessels, having high daily opera- tional costs. During the last twenty years, the research community has been making an effort to design new Intervention Autonomous Underwater Vehicles (I-AUV), which could, in the near future, replace the ROVs, significantly decreasing these costs. Until now, the experimental work using I-AUVs has been limited to a few single-vehicle interventions, including object search and recovery, valve turning, and hot stab operations. More complex scenarios usually require the cooperation of multiple agents, i.e., the transportation of large and heavy objects. Moreover, using small, autonomous vehicles requires consideration of their limited load capacity and limited manipulation force/torque capabilities. Following the idea of multi-agent systems, in this paper we propose a possible solution: using a group of cooperating I-AUVs, thus sharing the load and optimizing the stress exerted on the manipulators. Specifically, we tackle the problem of transporting a long pipe. The presented ideas are based on a decentralized Task-Priority kinematic control algorithm adapted for the highly limited communication bandwidth available underwater. The aforementioned pipe is transported following a sequence of poses. A path-following algorithm computes the desired velocities for the robots’ end-effectors, and the on-board controllers ensure tracking of these setpoints, taking into account the geometry of the pipe and the vehicles’ limitations. The utilized algorithms and their practical implementation are discussed in detail and validated through extensive simulations and experimental trials performed in a test tank using two 8 DOF I-AUVs This work was supported in part by the TWINBOT ’’TWIN ROBOTS FOR COOPERATIVE UNDERWATER INTERVENTION MISSIONS’’ Project funded by the Spanish Ministry of economy, industry, and competitiveness, under Grant DPI2017-86372-C3, in part by the Valencian Government (CIRTESU Project) under Grant IDIFEDER/2018/013, and in part by the Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement de la Generalitat de Catalunya under Grant 2019FI_B_00812

Being able to teach complex capabilities, such as folding garments, to a bi-manual robot is a very challenging task, which is often tackled using learning from demonstration datasets. The few garment folding datasets available nowadays to the robotics research community are either gathered from human demonstrations or generated through simulation. The former have the huge problem of perceiving human action and transferring it to the dynamic control of the robot, while the latter requires coding human motion into the simulator in open loop, resulting in far-from-realistic movements. In this article, we present a reduced but very accurate dataset of human cloth folding demonstrations. The dataset is collected through a novel virtual reality (VR) framework we propose, based on Unity¿s 3D platform and the use of a HTC Vive Pro system. The framework is capable of simulating very realistic garments while allowing users to interact with them, in real time, through handheld controllers. By doing so, and thanks to the immersive experience, our framework gets rid of the gap between the human and robot perception-action loop, while simplifying data capture and resulting in more realistic samples., This work was developed in the context of the project CLOTHILDE (”CLOTH manIpulation Learning from DEmonstrations”) which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No. 741930) and is also supported by the BURG project PCI2019-103447 funded by MCIN/ AEI /10.13039/501100011033 and by the ”European Union”.

A dia d’avui trobem intel·ligència artificial aplicada a gran diversitat de camps, la podem trobar als telèfons mòbils, presents a la butxaca de qualsevol persona fins als hospitals o als bancs. L’aplicació de la intel·ligència artificial amb combinació de la robòtica ha representat una millora de la qualitat de vida. Des-de feines tedioses i monòtones fins a tasques perilloses o complicades que abans es duien a terme per persones ara es fan amb robots. Els robots autònoms són capaços de reconèixer i aprendre sobre el seu entorn i prendre decisions de forma independent sense la necessitat de control humà. Varien amb dimensions, funcionalitat, mobilitat, etc. en funció de les seves aplicacions. Es poden trobar als habitatges com els robot Roomba, processos industrials o destinats a l’exploració. La percepció es molt important en el robots autònoms, ja que abans de prendre decisions han d’avaluar informació sobre l’entorn. L’objectiu d’aquest treball de final grau serà dotar un robot quadrúpede, el Bittle, amb percepció visual a traves de l’entrenament d’una xarxa neuronal que utilitza el model Mask RCNN. Controlat per una Raspberry Pi, es buscara que el Bittle sigui capaç de detectar pilotes i actuar amb conseqüència

Treballs recents sobre localització de vehicles i mapeig dels seus entorns es desenvolupen per a dispositius portàtils o robots terrestres que assumeixen moviments lents i suaus. Contràriament als entorns de curses d’alta velocitat. Aquesta tesi proposa un nou model d’SLAM, anomenat LIMO-Velo, capaç de corregir el seu estat amb una latència extremadament baixa tractant els punts LiDAR com un flux de dades. Els experiments mostren un salt en robustesa i en la qualitat del mapa mantenint el requisit de correr en temps real. El model aconsegueix una millora relativa del 20% en el KITTI dataset d’odometria respecte al millor rendiment existent; no deriva en un sol esce- nari. La qualitat del mapa a nivell de centı́metre es manté amb velocitats que poden arribar a 20 m/s i 500 graus/s. Utilitzant les biblioteques obertes IKFoM i ikd-Tree, el model funciona x10 més ràpid que la majoria de models d’última generació. Mostrem que LIMO-Velo es pot generalitzar per exe- cutar l’eliminació dinàmica d’objectes, com ara altres agents a la carretera, vianants i altres. Trabajos recientes sobre la localización de vehı́culos y el mapeo de sus en- tornos se desarrollan para dispositivos portátiles o robots terrestres que asumen movimientos lentos y suaves. Al contrario de los entornos de carreras de alta velocidad. Esta tesis propone un nuevo modelo SLAM, LIMO-Velo, capaz de corregir su estado en latencia extremadamente baja al tratar los puntos LiDAR como un flujo de datos. Los experimentos muestran un salto en la solidez y la calidad del mapa mientras se mantiene el requisito de tiempo real. El modelo logra una mejora relativa del 20% en el conjunto de datos de KITTI Odometry sobre el mejor desempeño existente; no deriva en un solo escenario. La calidad del mapa de nivel centimétrico todavı́a se logra a velocidades de carrera que pueden llegar hasta 20 m/s y 500 grados/s. Us- ando las bibliotecas abiertas IKFoM e ikd-Tree, el modelo funciona x10 más rápido que la mayorı́a de los modelos de última generación. Mostramos que LIMO-Velo se puede generalizar para trabajar bajo la eliminación dinámica de objetos, como otros agentes en la carretera, peatones y más. Recent works on localizing vehicles and mapping their environments are de- veloped for handheld devices or terrestrial robots which assume slow and smooth movements. Contrary to high-velocity racing environments. This thesis proposes a new SLAM model, LIMO-Velo, capable of correcting its state at extreme low-latency by treating LiDAR points as a data stream. Experiments show a jump in robustness and map quality while maintaining the real-time requirement. The model achieves a 20% relative improvement on the KITTI Odometry dataset over the existing best performer; it does not drift in a single scenario. Centimeter-level map quality is still achieved under racing velocities that can go up to 20m/s and 500deg/s. Using the IKFoM and ikd-Tree open libraries, the model performs x10 faster than most state-of-the-art models. We show that LIMO-Velo can be generalized to work under dynamic object removal such as other agents in the road, pedestrians, and more. Outgoing

Este proyecto ha tenido como objetivo principal la construcción de un robot capaz de jugar autónomamente al conocido juego 2048. Se puede desglosar el trabajo en dos grandes partes: una de software y una de hardware. En lo que respecta a la primera, se ha realizado un análisis inicial de diversos aspectos del juego y se ha comparado distintos métodos de resolución, principalmente minimax, expectimax y redes neuronales. De todos estos métodos se han propuesto multitud de variantes en las que se han añadido diversas características como, por ejemplo, las distintas heurísticas para evaluar los nodos del árbol de búsqueda en los métodos minimax y expectimax o distintas arquitecturas en el caso de las redes neuronales. En cuanto al hardware, se ha construido un robot ideado para ser posicionado delante de la pantalla de un ordenador con el juego abierto e interaccionar con el ordenador, dando las órdenes de movimiento a través del puerto USB. El robot incorpora una placa Raspberry Pi 4, una placa Pro Micro ATmega32u4, una cámara y componentes de soporte, todo ello, dentro de una carcasa hecha a partir de modelado e impresión 3D. The main objective of this project was the construction of a robot capable of autonomously playing the well-known game 2048. The work has two main parts: software and hardware. Regarding the first, an initial analysis of various aspects of the game has been carried out and different resolution methods have been compared, mainly minimax, expectimax and neural networks. Many variants of all these methods have been proposed in which various features have been added, such as, for example, the different heuristics to evaluate the nodes of the search tree in the minimax and expectimax methods or different architectures in the case of neural networks. For the hardware, a robot has been built and designed to be positioned in front of a computer screen with the game open and interact with the computer, giving movement orders through the USB port. The robot incorporates a Raspberry Pi 4 board, a Pro Micro ATmega32u4 board, a camera and supporting components, all inside a 3D modelled and printed case.

Robots are usually equipped with 3D range sensors such as laser line scanners (LLSs) or lidars. These sensors acquire a full 3D scan in a line by line manner while the robot is in motion. All the lines can be referred to a common coordinate frame using data from inertial sensors. However, errors from noisy inertial measurements and inaccuracies in the extrinsic parameters between the scanner and the robot frame are also projected onto the shared frame. This causes a deformation in the final scan containing all the lines, which is known as motion distortion. Rigid point cloud registration with methods like ICP is therefore not well suited for such distorted scans. In this paper we present a non-rigid registration method that finds the rigid transformation to be applied to each line in the scan in order to match an existing model. We fully leverage the continuous and relatively smooth robot motion with respect to the scanning time to formulate our method reducing the computational complexity while improving accuracy. We use synthetic and real data to benchmark our method against a state-of-the-art non-rigid registration method. Finally, the source code for the algorithm is made publicly available., IEEE Robotics and Automation Letters, 7 (3), ISSN:2377-3766

This article studies the design, modeling, and implementation of a path-following algorithm as a guidance, navigation, and control (GNC) architecture for an autonomous underwater vehicle (AUV). First, a mathematical model is developed based on nonlinear equations of motion and parameter estimation techniques, including the model validation based on field test data. Then, the guidance system incorporates a line-of-sight (LOS) algorithm with a combination of position PID controllers. The GNC architecture includes a modular and multi-layer approach with an LOS-based, path-following algorithm in the AUV platform. Furthermore, the navigation used in the path-following algorithm is developed based on a predefined coverage area. Finally, this study addresses simulation and field test control scenarios to verify the developed GNC architecture This research was funded by the project MOCAV-G500 from EUMarineRobots that has received funding from the European Union’s Horizon 2020 (grant agreement No 731103), with the collaboration of the aColor project, which is funded by Technology Industries of Finland Centennial and Jane & Aatos Erkko Foundations

Trabajo fin de máster presentado en la Universidad Politécnica de Cataluña, Máster en Automática y Robótica.--2021-10-08, A fundamental necessity in mobile robotics is the ability of knowing the robot own location. In many applications, a map of the environment is not available or it does not have the accuracy or the information necessary for robot localization. In this cases, it is necessary to build a map simultaneously (SLAM). In applications where the GPS signal is not reliable, like in indoor environments, the localization has to be based in estimating robot position with the information of the robot movement. This estimation is obtained in local coordinates and it implies that a drift in the estimation accumulates error over time. In order to reduce the drift effects, it is necessary the capacity to recognize places that had already been visited and readjust the localization. This search of loop closures provides global reference points to the localization system. This field of study is already a very discussed topic, but finding an equilibrium between efficiency and robustness in order to be able to work in large environments is still a challenge. The aim of this work is to equip the WOLF SLAM library with an state-of-art loop closure for 2D LIDAR sensors. WOLF SLAM library is integrated with ROS and proposes an interesting environment to manage SLAM and localization problems based in graph-SLAM. In this work, we first performed a review of the state of the art for loop closure that lead us to FALKO. It is an open source project focused in loop closure that implements feature detection and scene matching using descriptors. Furthermore, the FALKO loop closure algorithm has been integrated along with an Iterative Closest Point method for robustness and accuracy purposes. ICP is a point to line metric optimized for range-finder scan matching. The integration and developed algorithms has been tested with experimental data from a real robot

Aquest document presenta un mil·lirobot autònom que persegueix un camp magnètic. El robot és senzill i petit de 57 mm per 49 mm per 18,5 mm i és capaç de rastrejar el camp magnètic. Es presenta i s’explica com s’ha dissenyat i fabricat el robot, el seu control, el seu model dinàmic com l’agorisme elaborat i la presa de decisions autònomes. El robot té moltes aplicacions possibles, com ara el seguiment d’SPIONS al cervell en el pacients que tenen càncer cerebral This document presents an autonomous swimming millirobot driven by a magnetic field. The boat is simple and small (57mm by 49mm by 18.5mm) and it can track the magnetic field. The development of the hardware and algorithm is presented as its dynamic model, perception, and autonomous decision making. The robot has many possible applications, such as tracking SPIONS in the brain Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura

Des de fa uns anys s’han començat a investigar i a desenvolupar potes mecàniques dels autòmats, pels grans avantatges que comporta, tal com la seva facilitat d’adaptació per terrenys irregulars i capacitat de maniobra. Theo Jansen, enginyer per la universitat de Delft va causar sensació per la construcció de criatures autòmats les quals caminen per la platja a causa del vent. Aquesta aprofiten aquesta energia per moure les potes articulades fetes mitjançant tubs de PVC. En aquest treball he volgut reproduir l’anàlisis cinemàtic i dinàmic de Theo Jansen. Es descriu analíticament l’obtenció de les propietats cinemàtiques i dinàmiques per descriure el moviment de les potes. Així mateix s’ha fet ús de Matlab per comprovar la trajectòria del punt de contacte amb el terra de les potes i també del Solidworks per realitzar un anàlisis dels esforços i deformacions que rep aquesta. Amb L’objectiu final de poder imprimir per mitja d’una impressora 3D un robot quadrúpede amb les potes de Theo Jansen. Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura

This paper presents the development of unmanned marine robotic control modelling and control synthesis using a coupled multivariable underactuated nonlinear adaptive U-model approach. The proposed controller was developed using thru an open source robot operating system (ROS) platform. The new adaptive coupled U-model based internal model control (IMC) node was successfully developed and tested. The proposed controller demonstrated the simplicity of the control synthesis process and the implementation of the mathematical algorithm in real-time. The controller was compared with the proven existing GIRONA 500 UUV for real-time performance. The ROS environment provides fast and reliable controller design and development compared to conventional software architecture. Simulation and real-time experiment were conducted using ROS via the GIRONA 500 UUV platform and compared with a PID mission controller. A new ROS node of nonlinear adaptive U-model based IMC was developed using ROS. The results showed good control signal convergence and tracking performance between the plant or system model with the proposed method This research was supported by the Yayasan Universiti Teknologi Petronas-Fundamental Research Grant-Petroleum Resource Fund (YUTP-FRG) funded by PRF under the grant 0015LC0-235, Universiti Teknologi PETRONAS, Malaysia and King Abdulaziz University, Jeddah, Saudi Arabia under the grant 015ME0-066

With the arrival of Industry 4.0, traditional manufactures will undergo a digitalization and automatization process to become smart factories. In such warehouses, fleets of fully autonomous robots will cooperate to transport required items from one place to another. In this context, it is important to guarantee that robots follow optimal collision-free trajectories. For this project, a setting with two KMR iiwa robots in a warehouse crossing is considered. These robots are composed of two main parts: the LBR iiwa (a mechanical arm) and the KMP iiwa (a holonomic mobile platform). Algorithms to control the KMP iiwa have been designed and implemented in ROS, and then tested using the simulator Gazebo. Starting from an already existing ROS-Gazebo model of the KMR iiwa, a closed-loop control system for navigation has been set up. Then, a navigation algorithm (to drive a single KMR iiwa robot by giving it a set of way-points) has been implemented. Afterwards, a collision avoidance strategy was added to the navigation algorithm. The designed policy took into account noisy data from laser scanners incorporated in the robot. To test the proposed strategy, several experiments have been performed considering various scenarios. The outcomes of these tests are analysed in this report. Finally, some conclusions and possible future work are presented. Con la llegada de la Industria 4.0, las fábricas tradicionales se verán sometidas a un proceso de digitalización i automatización para convertir-se en fábricas inteligentes. En estas manufacturas, flotas de robots autónomos colaboraran para transportar artículos necesarios de un punto a otro. En este contexto, es importante garantizar que los robots seguirán trayectorias óptimas, sin sufrir colisiones. En este proyecto, se ha considerado la situación de dos robots KMR iiwa navegando en una fábrica, dirigiéndose al mismo cruce de pasillos. Este tipo de robots está formado por: un LBR iiwa (un brazo mecánico) i un KMP iiwa (una plataforma holonòmica). En este trabajo, se han diseñado e implementado algoritmos de control para la KMP iiwa, utilizando ROS. El código desarrollado ha sido testado con el simulador Gazebo. Partiendo de un modelo ROS-Gazebo ya existente del robot KMR iiwa, un sistema de control en lazo cerrado ha sido implementado. Después, se ha desarrollado un algoritmo de navegación, con el que se puede conducir el robot KMR iiwa pasándole varios puntos de referencia del entorno. A continuación, se ha añadido una estrategia para evitar colisiones entre robots al algoritmo de navegación. El método diseñado tiene en consideración la información (con ruido) procedente de los escáneres laser con que el robot está equipado. Para testear la estrategia propuesta, se han llevado a cabo varios experimentos en distintos escenarios. El resultado de estos experimentos es analizado en este documento. Finalmente, se exponen las conclusiones del trabajo i posibles mejoras. Amb l'arribada de la Indústria 4.0, les fàbriques tradicionals es veuran sotmeses a un procés de digitalització i automatització per esdevenir fàbriques intel·ligents. En aquestes manufactures, flotes de robots autònoms cooperaran per transportar articles necessaris d'un lloc a un altre. En aquest context, és important garantir que els robots seguiran trajectòries òptimes, sense patir cap col·lisió. En aquest projecte, s'ha considerat la situació de dos robots KMR iiwa navegant en una fàbrica, dirigint-se al mateix encreuament de passadissos. Aquest tipus de robots estan formats per dues parts: un LBR iiwa (un braç mecànic) i un KMP iiwa (una plataforma holonòmica). En aquest treball, s'han dissenyat i implementat algoritmes de control per la KMP iiwa, fent servir ROS. El codi desenvolupat ha estat testat amb el simulador Gazebo. Partint d'un model ROS-Gazebo ja existent del robot KMR iiwa, un sistema de control en llaç tancat ha estat implementat. Després, s'ha desenvolupat un algoritme de navegació, amb el que es pot conduir el robot KMR iiwa passant-li un seguit de punts de referència del entorn. A continuació, s'ha afegit una estratègia per evitar col·lisions entre robots al algoritme de navegació. El mètode dissenyat té en compte la informació (amb soroll) proveint dels escàners làser amb els que estava equipat el robot. Per testar la estratègia proposada, s'han dut a terme varis experiments en diferents escenaris. El resultat d'aquests experiments son presentats en aquest document. Finalment, s'exposen les conclusions del treball i possibles millores.

L’objectiu d’aquest projecte és crear l’algoritme de control d’un robot de desinfecció UV-C, de manera que sigui el més eficient possible. Per una banda, s’ha dissenyat el comportament del robot utilitzant l’algoritme de pathfinding, o cerca de camí, A*, necessari per evitar la col·lisió amb obstacles o parets i per buscar la manera en què es porta a terme moviments complexos. També s’ha tingut en compte les característiques tècniques del robot; com per exemple, les dimensions de la base. Per altra banda, s’ha creat una interfície gràfica utilitzant Unity, de manera que en tot moment es pugui observar de forma senzilla el comportament de l’algoritme i el robot. El projecte s’ha estructurat en 3 fases, de manera que en cada una d’elles, l'algoritme pren més complexitat. En la primera fase s’ha creat la interfície i s’han programat les bases de l’algoritme, en la segona, s’ha tingut en compte la bateria, la càrrega i l’existència de varies sales i en la tercera, s’ha adjudicat un nivell de brutícia a les cel·les de les sales. Finalment, s’ha avaluat el rendiment de l’algoritme per cada una de les fases, concloent en quina de les casuístiques dels escenaris el seu comportament és més òptim. The objective of this project is to create the control algorithm of a UV-C disinfection robot so that it as efficiently as possible. On one hand, the robot performance has been designed using the pathfinding algorithm A*, useful to avoid colliding with obstacles or walls and to find how to make complex movements. Also, robot technical specifications have been taken into consideration; for example, the base dimensions. On another hand, a graphic interface has been created using Unity, so that at any moment the behaviour of the robot and algorithm can be analyzed easily. The project is structured in 3 phases, so in each one, the algorithm becomes more complex. In the first phase, the interface has been created and the bases of the algorithm have been programmed. In the second one, I considered the battery, the charging process and the possibility to have more than one zone. In the third one, each cell has a level of dirtiness attributed. Finally, in each phase, the performance has been evaluated, which allowed me to conclude in which cases the algorithm behaviour was more optimal.

En el present document es recull una primera fase de disseny, construcció i programació d'un robot mòbil autònom. Per a la lògica de control i el seu control remot s'ha emprat una Raspberry Pi 4B, la qual ha estat programada amb el llenguatge de programació Python. Amb aquest software s'han controlat els motors de corrent continu de 3V, que han fet avançar el robot, per un camí marcat amb línies negres, amb punts de referència obtinguts a través de tecnologia RFID. El robot és capaç d'esquivar objectes i interactuar amb l'entorn, amb la finalitat de donar ajuda en el servei de taula a la restauració, de manera que mitjançant una aplicació, el robot tingui capacitat de fer arribar el menjar als comensals. El prototip final se li han integrat diferents instrumentes de mesura i s'han testejat per l'obtenció d'uns primers resultats. Finalment s'han cercat informacions de les actuals tecnologies, per la proposta de possibles millores per la implementació final i real del projecte. En el presente documento se recoge una primera fase de diseño, construcción y programación de un robot móvil autónomo. Para la lógica de control y su control remoto se ha empleado una Raspberry Pi 4B, la cual ha sido programada con el lenguaje de programación Python. Con este software se han controlado los motores de corriente continua de 3V, que han hecho avanzar el robot, por un camino marcado con líneas negras, con puntos de referencia obtenidos a través de tecnología RFID. El robot es capaz de esquivar objetos e interactuar con el entorno, con el fin de dar ayuda en el servicio de mesa a la restauración, por lo que, mediante una aplicación, el robot tenga capacidad de hacer llegar la comida a los comensales. El prototipo final se le han integrado diferentes instrumentos de medida y se han testeado para la obtención de unos primeros resultados. Finalmente se han buscado informaciones de las actuales tecnologías, por la propuesta de posibles mejoras para la implementación final y real del proyecto. This document contains a first phase of design, construction and programming of an autonomous mobile robot. For the control logic and its remote control, a Raspberry Pi 4B has been used, which has been programmed with the Python programming language. With this software, the 3V direct current motors have been controlled, which have made the robot advance, along a path marked with black lines, with reference points obtained through RFID technology. The robot is capable of dodging objects and interacting with the environment, in order to provide help in catering for table service, so that through an application, the robot has the ability to deliver food to diners. Different measuring instruments have been integrated into the final prototype and they have been tested to obtain first results. Finally, information on current technologies has been sought, for the proposal of possible improvements for the final and real implementation of the project.

This paper presents a comparative study of docking algorithms intended for non-holonomic autonomous underwater vehicles, docking in funnel-shaped docking stations, operating under the influence of ocean currents. While descriptive surveys have been already reported in the literature, our goal is to compare the most relevant algorithms through realistic Monte Carlo simulations to provide an insight into their performance. To this aim, a new numerical performance indicator is proposed, which, based on the geometry of the manoeuvre, is able to characterize a successful or unsuccessful docking, providing a metric for comparison. The experimental study is carried out using hardware-in-the-loop simulation by means of the Stonefish simulator, including the dynamic/hydrodynamic model of the Sparus II AUV, models of all internal and external sensors, and the collision geometry representing the docking station This work was supported by the European Union’s Horizon 2020 Research and Innovation Program through the ATLANTIS ‘‘The Atlantic Testing Platform for Maritime Robotics: New Frontiers for Inspection and Maintenance of Offshore Energy Infrastructures’’ Project under Grant 871571

In this paper, we compare liability rules in a world where human-driven and fully-autonomous cars coexist. We develop a model where a manufacturer can invest to improve the safety of autonomous cars. Human drivers may decide to purchase a fully-autonomous car to save precaution costs to avoid road accidents and shift liability to the car manufacturer. As compared to the negligence rule, a strict liability regime on both human drivers and car manufacturers is proved to be a superior policy. In particular, strict liability leads to more efficient R&D investments to enhance the benefits of the technology and favors the adoption of fully-autonomous cars. We also recommend that users of fully-autonomous cars make a technology-dependent payment to a third-party if there is an accident to discipline their activity levels.

This paper proposes a 3D object recognition method for non-coloured point clouds using point features. The method is intended for application scenarios such as Inspection, Maintenance and Repair (IMR) of industrial sub-sea structures composed of pipes and connecting objects (such as valves, elbows and R-Tee connectors). The recognition algorithm uses a database of partial views of the objects, stored as point clouds, which is available a priori. The recognition pipeline has 5 stages: (1) Plane segmentation, (2) Pipe detection, (3) Semantic Object-segmentation and detection, (4) Feature based Object Recognition and (5) Bayesian estimation. To apply the Bayesian estimation, an object tracking method based on a new Interdistance Joint Compatibility Branch and Bound (IJCBB) algorithm is proposed. The paper studies the recognition performance depending on: (1) the point feature descriptor used, (2) the use (or not) of Bayesian estimation and (3) the inclusion of semantic information about the objects connections. The methods are tested using an experimental dataset containing laser scans and Autonomous Underwater Vehicle (AUV) navigation data. The best results are obtained using the Clustered Viewpoint Feature Histogram (CVFH) descriptor, achieving recognition rates of 51.2%, 68.6% and 90%, respectively, clearly showing the advantages of using the Bayesian estimation (18% increase) and the inclusion of semantic information (21% further increase).

With the arrival of Industry 4.0, traditional manufactures will undergo a digitalization and automatization process to become smart factories. In such warehouses, fleets of fully autonomous robots will cooperate to transport required items from one place to another. In this context, it is important to guarantee that robots follow optimal collision-free trajectories. For this project, a setting with two KMR iiwa robots in a warehouse crossing is considered. These robots are composed of two main parts: the LBR iiwa (a mechanical arm) and the KMP iiwa (a holonomic mobile platform). Algorithms to control the KMP iiwa have been designed and implemented in ROS, and then tested using the simulator Gazebo. Starting from an already existing ROS-Gazebo model of the KMR iiwa, a closed-loop control system for navigation has been set up. Then, a navigation algorithm (to drive a single KMR iiwa robot by giving it a set of way-points) has been implemented. Afterwards, a collision avoidance strategy was added to the navigation algorithm. The designed policy took into account noisy data from laser scanners incorporated in the robot. To test the proposed strategy, several experiments have been performed considering various scenarios. The outcomes of these tests are analysed in this report. Finally, some conclusions and possible future work are presented. Con la llegada de la Industria 4.0, las fábricas tradicionales se verán sometidas a un proceso de digitalización i automatización para convertir-se en fábricas inteligentes. En estas manufacturas, flotas de robots autónomos colaboraran para transportar artículos necesarios de un punto a otro. En este contexto, es importante garantizar que los robots seguirán trayectorias óptimas, sin sufrir colisiones. En este proyecto, se ha considerado la situación de dos robots KMR iiwa navegando en una fábrica, dirigiéndose al mismo cruce de pasillos. Este tipo de robots está formado por: un LBR iiwa (un brazo mecánico) i un KMP iiwa (una plataforma holonòmica). En este trabajo, se han diseñado e implementado algoritmos de control para la KMP iiwa, utilizando ROS. El código desarrollado ha sido testado con el simulador Gazebo. Partiendo de un modelo ROS-Gazebo ya existente del robot KMR iiwa, un sistema de control en lazo cerrado ha sido implementado. Después, se ha desarrollado un algoritmo de navegación, con el que se puede conducir el robot KMR iiwa pasándole varios puntos de referencia del entorno. A continuación, se ha añadido una estrategia para evitar colisiones entre robots al algoritmo de navegación. El método diseñado tiene en consideración la información (con ruido) procedente de los escáneres laser con que el robot está equipado. Para testear la estrategia propuesta, se han llevado a cabo varios experimentos en distintos escenarios. El resultado de estos experimentos es analizado en este documento. Finalmente, se exponen las conclusiones del trabajo i posibles mejoras. Amb l'arribada de la Indústria 4.0, les fàbriques tradicionals es veuran sotmeses a un procés de digitalització i automatització per esdevenir fàbriques intel·ligents. En aquestes manufactures, flotes de robots autònoms cooperaran per transportar articles necessaris d'un lloc a un altre. En aquest context, és important garantir que els robots seguiran trajectòries òptimes, sense patir cap col·lisió. En aquest projecte, s'ha considerat la situació de dos robots KMR iiwa navegant en una fàbrica, dirigint-se al mateix encreuament de passadissos. Aquest tipus de robots estan formats per dues parts: un LBR iiwa (un braç mecànic) i un KMP iiwa (una plataforma holonòmica). En aquest treball, s'han dissenyat i implementat algoritmes de control per la KMP iiwa, fent servir ROS. El codi desenvolupat ha estat testat amb el simulador Gazebo. Partint d'un model ROS-Gazebo ja existent del robot KMR iiwa, un sistema de control en llaç tancat ha estat implementat. Després, s'ha desenvolupat un algoritme de navegació, amb el que es pot conduir el robot KMR iiwa passant-li un seguit de punts de referència del entorn. A continuació, s'ha afegit una estratègia per evitar col·lisions entre robots al algoritme de navegació. El mètode dissenyat té en compte la informació (amb soroll) proveint dels escàners làser amb els que estava equipat el robot. Per testar la estratègia proposada, s'han dut a terme varis experiments en diferents escenaris. El resultat d'aquests experiments son presentats en aquest document. Finalment, s'exposen les conclusions del treball i possibles millores.

La realització d’aquest treball de final de grau consisteix en l’estudi i la programació d’un algoritme de seguiment de carreteres en un vehicle autònom. Avui dia, un dels propòsits per les empreses automobilístiques és aconseguir desenvolupar diferents algoritmes per l’automatització total del vehicle, mitjançant la incorporació de sensors, i així, oferir comoditat i seguretat als seus passatgers. Però abans de la posada en marxa en automòbils cal fer proves per desenvolupar i verificar el seu funcionament. En l’estudi, utilitzarem un vehicle a escala reduïda, que incorpora diferents tipus de sensors per aconseguir la seva autonomia, en aquest projecte es farà mitjançant el tractament d’imatges proporciones per una càmera digital. El contingut d’aquest projecte, consisteix a donar els fonaments teòrics i pràctics per poder crear un node de ROS perquè el vehicle pugui recórrer autònomament una carretera. En el que s’explicaran les diferents funcions per l’adquisició i el tractament de la imatge utilitzant OpenCV, i així, escollir la millor opció per percebre les línies delimitadores de les carreteres. I per tant, resoldre els diferents casos perquè el vehicle no surti del trajecte, n’hi ha dos de rellevants: quan visualitzem els dos costats de la carretera i quan només visualitzem un costat. Finalment, aconseguir que generi els punts que marquin el seguiment de la trajectòria i publicar-los al node que acciona la transmissió i la direcció del moviment del cotxe. The realisation of this final project consists at the study and the programming of an algorithm of follow-up of roads in an autonomous vehicle. Nowadays, one of the purposes for the car companies is to develop different algorithms for the total automation of the vehicle, by means of the incorporation of sensors, and offer comfort and security at his passengers. But before the set up at cars it is necessary to make tests to develop and verify his operation. At the study, will utilise a vehicle a scale reduced, that incorporates distinct type of sensors to achieve his autonomy, at this project will make by means of the treatment of images proportioned by the digital camera. The content of this project consists at giving the theoretical and practical foundations for creating a ROS’s node so that the vehicle can independently cross a road. For explanations of the different functions for image acquisition and processing, use OpenCV and thus choose the best option for perceiving the roads. In the objective of reselling the different cases so that the car does not leave the route, there are two of notable: when we visualize both sides of the highway and when we only visualize one side. Finally, it is advisable to generate the points that mark the track of the trajectory and publish them to the node that drives the transmission and direction of the car’s movement.

In this field note, we detail the operations and discuss the results of an experiment conducted in the unstructured environment of an underwater cave complex using an autonomous underwater vehicle (AUV). For this experiment, the AUV was equipped with two acoustic sonar sensors to simultaneously map the caves' horizontal and vertical surfaces. Although the caves' spatial complexity required AUV guidance by a diver, this field deployment successfully demonstrates a scan‐matching algorithm in a simultaneous localization and mapping framework that significantly reduces and bounds the localization error for fully autonomous navigation. These methods are generalizable for AUV exploration in confined underwater environments where surfacing or predeployment of localization equipment is not feasible, and they may provide a useful step toward AUV utilization as a response tool in confined underwater disaster areas This research work was partially sponsored by the EU FP7-Projects: Tecniospring- Marie Curie (TECSPR13-1-0052), MORPH (FP7-ICT-2011-7-288704), Eurofleets2 (FP7-INF-2012-312762), and the National Science Foundation (OCE-0955674).

Durante los últimos años, uno de los productos más vendidos durante las jornadas de rebajas conocidas como Black Friday o Cyber Monday ha sido el robot aspirador. Sus avanzadas prestaciones, que permiten su puesta en marcha y funcionamiento cuando el usuario está ausente de su domicilio, han seducido de forma clara a los compradores. Pero este éxito de ventas ha venido acompañado también de algunas dudas: ¿están estos robots adquiriendo datos sensibles que pueden ser vendidos en el mercado global de los datos personales?

Nowadays, autonomous robots for all different purposes are the state-of-the-art technology, all companies across the industry are implementing all kind of autonomous machines to control and computerize the industrial systems and processes, making errors virtually impossible and, despite all the first investment, the transition pays off in the mid and long-term because of all the time and resources saved. This new industrial phase, which is a current tendency commonly known as Industry 4.01 , focuses on interconnectivity, automation, machine learning, and real-time data. The same trend applies to vehicles or mobile machines (for industrial or non-industrial purposes), the use of fossil fuels is decreasing whereas the implementation of automation and systems of artificial intelligence is strongly increasing day by day. Heavy-duty hydraulically driven mobile working machines (for mining, construction, forestry, and material handling purposes) are also in constant development for increasing efficiency and performance because of the higher requirements set by the constant growth of this sector. As a result, working machines are increasingly expected to become more precise, and robotics like towards autonomous operation. This thesis studies the design requirements and sizing for each hydraulic actuator system of the case of study, a long-reach manipulator 4x4 Haulotte HA16 RTJ PRO which is available at the University of Tampere (Hervanta Campus). This hydraulic articulated boom, originally manually driven, is being redesigned into a servo-hydraulic system maintaining the original diesel motor source and taking into account its limited power. The general work consists of requirements of development, hydraulic engineering designs, component selection, modeling, and simulations towards a new prototyped system meeting the set requirements. During the making of this thesis, the component selection was already done and the scope of the study relies on the kinematic and hydraulic analysis for the vehicle. Also, this thesis has pursued another objective, which is the compilation of technical data (such as the boom dimensions, original hydraulics, or technical information about the main components) among the different reports about the Haulotte, trying to achieve that the upcoming students or researchers can get clearer ideas and the maximum possible information about the case of study reading this document. The goals are got from a theoretical viewpoint through equations and hypotheses considered throughout the study, therefore, the results obtained must be taken as approximations. The reason is the coronavirus pandemic, which forced the university and the laboratories to close since the beginning of the semester and some parameters or variables taken may not always coincide with the reality because were taken from a 3D model from the original Haulotte website. Keywords: Wheeled Mobile Robot, Mobile platform, Hydraulics, Automation, Kinematics, Haulotte, HA16 RTJ PRO, Boom, Articulating boom lift, 4WSD

L’empresa de robòtica mòbil MiR està introduïnt al mercat els robots AMR (Autonomous Mobile Robots), els quals es caracteritzen pel fet de no ser guiats. És a dir, aquesta tipologia de robot no necessita de cap ajuda externa per al seu moviment, fet pel qual és necessaria una bona detecció dels objectes estàtics i dinàmics per evitar colisions. Des de MiR es vol desenvolupar un dispositiu que permeti el control de tràfic de les fàbriques a partir d’una comunicació remota. L’objectiu del disseny d’aquest dispositiu és poder saber la posició d’un objecte dins la fàbrica i comunicar-se amb els robots. En funció de l’existència d’algun impediment, aquests podran passar, o no, per determinats llocs, entre altres. El present projecte consisteix en la programació d’un dispositiu basat en una Raspberry Pi per al càlcul del seu posicionament en interiors i la comunicació de forma remota amb la flota de robots MiR; el sistema de gestió de robots pròpi de MiR. Per aquest motiu, en el codi desenvolupat s’hi troba el càlcul del posicionament en interiors a partir de la tirangulació de WIFI i els codis per a la comunicació remota per a obtenir els mapes dels robots i les zones d’aquests. També s’hi troba la verificació de si l’objecte que té el dispositiu es troba dins d’una zona de límit de robots, i en el cas que s’hi trobi, es bloquegi la zona per tal que cap robot pugui entrar-hi. D’aquesta manera, s’evita que el robot arribi a mostrar un error per no poder passar per un determinat lloc, i també s’eviten possibles col·lisions. La empresa de robótica móbil MiR está introduciendo en el mercado los robots AMR (Autonomous Mobile Robots), los cuales se caracterizan por el hecho de no ser guiados. Es decir, esta tipología de robots no necesita de ninguna ayuda externa para su movimiento, por lo que es necesaría una buena deteccón de los objetos estáticos y dinámicos para evitar colisiones. Des de MiR se quiere desarrollar un dispositivo que permita el control del tráfico de la fábricas a partir de una comunicación remota. El objetivo del diseño de este dispositivo es poder saber la posición de un objeto dentro de la fábrica y comunicar-se con los robots. En función de la existencia de algun impedimiento, éstos podran pasar, o no, por determinados sitios, entre otras cosas. El presente proyecto consiste en la programación de un dispositivo basado en una Raspberry Pi para el cálculo de su posicionamiento en interiores y la comunicación de forma remota con una flota de robots MiR; el sistema de gestión de robots propio de MiR. Por este motivo, en el código desarrollado se encuentra el cálculo del posicionamiento en interiores a partir de la triangulación de WIFI y los códigos para la comunicación remota para obtener los mapas de los robots y las zonas de estos. También se encuentra la verificación de si un objeto que posee el dispositivo se encuentra dentro de una zona de límite de robots, y en el caso que se encuentre, se bloquee la zona para que ningún robot pueda entrar. De esta forma, se evita que el robot llegue a enseñar un error por no poder pasar por un determinado sitio, y también se evitan posibles colisiones. The mobile robotics company MiR is introducing the AMR (Autonomous Mobile Robots) robots in the market, which are characterized by not being guided. This means, that this type of robot needs not any external help to move, this is why a good dynamic and static objects detection is necessary to avoid collisions. From MiR, we want to develop a device that allows the control of factory traffic using remote communication. The objective of the device design is to be able to know an object position in a factory and communicate with robots. Depending on the existence of some impediment, these may or may not pass through certain sites, among other things. The current project consists of the programming of a device based on a Raspberry Pi for the calculation of its indoor positioning and remote communication with a MiR robots Fleet; MiR’s own robot management system. For this reason, the code developed includes the calculation of the indoor positioning from the WIFI triangulation and the codes for the remote communication to get the robot maps and their zones. There is also the verification of whether an object held by the device is in a limit robots zone, and if it is, the zone is blocked so that no robot can enter. In this way, the robot is prevented from showing an error because it cannot pass through a certain place, and possible collisions are also avoided.

One of the problems that service robotics deals with is to bring mobile manipulators to work in semi-structured human scenarios, which requires an efficient and flexible way to execute every-day tasks, like serve a cup in a cluttered environment. Usually, for those tasks, the combination of symbolic and geometric levels of planning is necessary, as well as the integration of perception models with knowledge to guide both planning levels, resulting in a sequence of actions or skills which, according to the current knowledge of the world, may be executed. This paper proposes a planning and execution framework, called SkillMaN, for robotic manipulation tasks, which is equipped with a module with experiential knowledge (learned from its experience or given by the user) on how to execute a set of skills, like pick-up, put-down or open a drawer, using workflows as well as robot trajectories. The framework also contains an execution assistant with geometric tools and reasoning capabilities to manage how to actually execute the sequence of motions to perform a manipulation task (which are forwarded to the executor module), as well as the capacity to store the relevant information to the experiential knowledge for further usage, and the capacity to interpret the actual perceived situation (in case the preconditions of an action do not hold) and to feed back the updated state to the planner to resume from there, allowing the robot to adapt to non-expected situations. To evaluate the viability of the proposed framework, an experiment has been proposed involving different skills performed with various types of objects in different scene contexts

Encoding practical knowledge about everyday activities has proven difficult, and is a limiting factor in the progress of autonomous robotics. Learning approaches, e.g. imitation learning from human data, have been used as a way to circumvent this difficulty. While such approaches are on the right track, they require comprehensive knowledge modelling about the data present in records of activity episodes, and about the skills one attempts to have the robot learn. We provide a list of competency questions such knowledge modelling should answer, summarize some recent developments in this direction, and finish with a few open problems.

The increment of dependant people for the realization of Activities of Daily Living is a fact that comes implicit with the increase of population and their lifespan given by the modern technological progress. For this reason, Assistive Robotics is a field that is currently researched and gradually implemented. In this project, two different algorithms to train robots to perform those tasks are being studied and compared, Reinforcement Learning and Imitation Learning. The main objective is to learn the benefits, drawbacks and suitability of both for a given task. To do the experimentation, the Assistive Gym software will be used, to train the nets with Reinforcement Learning and to perform the movements with Dynamic Movement Primitives learned from a demonstration. The environment will consist in a Sawyer robot cleaning the right arm of a human lying down on a bed. The potential of each methodology will be tested in different scenarios, having the user’s arm straight or partially flexed. Furthermore, small implementations to improve the execution of the task will be implemented for the Dynamic Movement Primitives to test the performance and the similarity to the demonstration. Both methodologies are capable to perform the desired task successfully, the model trained with Reinforcement Learning is more suitable for an individual user that will use the robot in a long term, due to the training time. The Imitation Learning algorithm is faster to teach and its trajectory and velocities are easily adaptable to the preferences of the user, it is more suitable in places where there is a high rotation of users that require urgent and short attention, such as clinics or hospitals.

En aquest treball de fi de grau es realitza l’habilitació d’un prototip no documentat anomenat Dumbo. Per aconseguir aquest propòsit s’inicia desmuntant-lo per tal de deixar únicament la part de la plataforma mòbil i l’electrònica, elements els quals seran els reaprofitats. El primer pas d’aquest projecte és la identificació del hardware que incorpora aquest prototip, a continuació, l’estudi dels mateixos i del seu connexionat i, concloent l’objectiu principal, la interpretació del software que incorpora. Establertes i definides les funcionalitats que incorpora aquesta plataforma mòbil es deixa habilitada perquè en treballs futurs se li puguin afegir capacitats que la millorin, tot portant-la cap a un robot mòbil autònom. Un cop concretat l’estat de la plataforma mòbil i habilitada, s’estudia el disseny d’un sistema de percepció de l’entorn mitjançant el sensor RPLidar A2 (aportant la mesura de les distàncies dels objectes sòlids que l’envolten) integrat amb la placa Raspberry Pi 4 (aportant la gestió de les mesures realitzades pel sensor), guiant la integració d’aquests components a la plataforma mòbil sota estudi. Recopilant els objectius d’aquest projecte resulta un primer pas cap a la conversió d’aquest prototip cap a un robot mòbil autònom possiblement intel·ligent.

L’aplicació de ROS (Robot Operating System) està cada vegada més present tant en el món laboral com en el competitiu. Amb aquest treball es pretén introduir al lector en aquest aspecte, portant a terme una aplicació basada en ROS dins l’àmbit d’una competició de robòtica de nivell Europeu (Eurobot).PUCRA, l’associació de robòtica d’estudiants de l’EEBE, ha sigut des d’on ha partit aquest projecte. En cert punt aquest ha pres una altre direcció respecte a la marcada des de l’associació. Així doncs l’adaptabilitat d’un per canvis en aquest sentit ha sigut clau per prosseguir endavant.Partint de la base d’un packageexistent, l’objectiu es modificar-lo per adaptar-lo als requeriments desitjats en condició del reglament de la competició, com a objected’estudi el càlcul de trajectòries. Mitjançant un protocol de proves es porta un registre dels canvis realitzats pas a pas fins a arribar a la solució final. La qual pren en principal importància l’optimització dels moviments del robot dins l’espai de la competició, concloent amb el fet de que Dijkstra resulta ser l’algoritme, tot i que el més antic, el més eficaç donades les prestacions que ROS atorga per defectei donats els resultats de les proves realitzades.

Autonomous indoor service robots are supposed to accomplish tasks, like serve a cup, which involve manipulation actions. Particularly, for complex manipulation tasks which are subject to geometric constraints, spatial information and a rich semantic knowledge about objects, types, and functionality are required, together with the way in which these objects can be manipulated. In this line, this paper presents an ontological-based reasoning framework called Perception and Manipulation Knowledge (PMK) that includes: (1) the modeling of the environment in a standardized way to provide common vocabularies for information exchange in human-robot or robot-robot collaboration, (2) a sensory module to perceive the objects in the environment and assert the ontological knowledge, (3) an evaluation-based analysis of the situation of the objects in the environment, in order to enhance the planning of manipulation tasks. The paper describes the concepts and the implementation of PMK, and presents an example demonstrating the range of information the framework can provide for autonomous robots.

We present an approach to endow an autonomous underwater vehicle with the capabilities to move through unexplored environments. To do so, we propose a computational framework for planning feasible and safe paths. The framework allows the vehicle to incrementally build a map of the surroundings, while simultaneously (re)planning a feasible path to a specified goal. To accomplish this, the framework considers motion constraints to plan feasible 3D paths, that is, those that meet the vehicle’s motion capabilities. It also incorporates a risk function to avoid navigating close to nearby obstacles. Furthermore, the framework makes use of two strategies to ensure meeting online computation limitations. The first one is to reuse the last best known solution to eliminate time-consuming pruning routines. The second one is to opportunistically check the states’ risk of collision. To evaluate the proposed approach, we use the Sparus II performing autonomous missions in different real-world scenarios. These experiments consist of simulated and in-water trials for different tasks. The conducted tasks include the exploration of challenging scenarios such as artificial marine structures, natural marine structures, and confined natural environments. All these applications allow us to extensively prove the efficacy of the presented approach, not only for constant-depth missions (2D), but, more important, for situations in which the vehicle must vary its depth (3D) J.D. Hern´andez, E. Vidal, N. Palomeras, and M. Carreras have been supported by the EXCELLABUST and ARCHROV Projects under the Grant agreements H2020-283TWINN-2015, CSA, ID: 691980 and DPI2014- 57746-C3-3-R respectively

Nowadays, it is quite common to have one unmanned aerial vehicle (UAV) working on a task but having a team of UAVs is still rare. One of the problems that prevent us from using teams of UAVs more frequently is flight planning. In this work, we present the first open-source solution ( https://pypi.org/project/pode/ ) for splitting any complex area into multiple parts. The area of interest can be convex or nonconvex and can include any number of no-flight zones. Four solutions, based on the algorithm of Hert and Lumelsky, are tested with the aim of improving the compactness of the partitions. We also show how the shape of the partitions influences flight performance in a real case scenario. This work was supported by Ministerio de Economía, Industria y Competitividad, and Gobierno de España under grants/award numbers BES-2017-079798 and TRA2016-77012-R.

The goal of the TFG would be to implement an object recognition system for an autonomous robot within the GRASS (Ground Autonomous Student Station) project. The idea is to use two cameras and be able to juxtapose the images and create a 3D vision of what the system is seeing (in our case the rover), and be able to differentiate different predetermined objects by the shape, and if it's possible by the color. The purpose consists in recognizing the position and the orientation of these objects with respect to the rover as the considered reference system. The GRound Autonomous Student Station (GRASS) is a student organization with the objective of building an autonomous rover, and one of the capacities needed is the possibility of identifying objects. This project is focused on creating a recognition system capable of being incorporated on the Rover. The developed recognition system is focused on detection by colour and shape, using four colours and one predetermined cylindrical shape. The system has to identify the distance between the object and the camera. The developed system includes a Raspberry Pi system with his camera, using Python and OpenCV libraries as software. The developed system has been tried in a real model, but the final implementation has not been ended due to reasons related with the GRASS project. The structural and logistic parts of the project are developing a new space distribution in order to include eciently the recognition system, although the system has been tested. The developed system is capable of detecting determined objects with a simple shape in a maximum range of 12 meters. At the same time, it works on real environments, counting illumination and environmental visual noise.

Action recognition is a challenging task that plays an important role in many robotic systems, which highly depend on visual input feeds. However, due to privacy concerns, it is important to find a method which can recognise actions without using visual feed. In this paper, we propose a concept for detecting actions while preserving the test subject&rsquo, s privacy. Our proposed method relies only on recording the temporal evolution of light pulses scattered back from the scene. Such data trace to record one action contains a sequence of one-dimensional arrays of voltage values acquired by a single-pixel detector at 1 GHz repetition rate. Information about both the distance to the object and its shape are embedded in the traces. We apply machine learning in the form of recurrent neural networks for data analysis and demonstrate successful action recognition. The experimental results show that our proposed method could achieve on average 96.47 % accuracy on the actions walking forward, walking backwards, sitting down, standing up and waving hand, using recurrent neural network.

EU Marine Robots project grant agreement no. 731103 The marine robotics community is lacking a high quality simulator for doing scientific research, especially when it comes to testing control and vision algorithms in realistic underwater intervention tasks. All of the solutions used today are either outdated or try to combine different software tools, which often results in bad performance, stability issues and lack of important features. This paper presents a new software tool, focused on, but not limited to, simulation of intervention autonomous underwater vehicles (I-AUV). It delivers advanced hydrodynamics based on actual geometry, simulation of underwater sensors and actuators, as well as realistic rendering of underwater environment and ocean surface. It consists of a library written in C++ and a Robot Operating System (ROS) package Patryk Ciésslak has received funding from the European Community H2020 Programme under the Marie Skłodowska-Curie grant agreement no. 750063 and under EU Marine Robots project grant agreement no. 731103

This paper describes a pose-based underwater 3D Simultaneous Localization and Mapping (SLAM) using a multibeam echosounder to produce high consistency underwater maps. The proposed algorithm compounds swath profiles of the seafloor with dead reckoning localization to build surface patches (i.e., point clouds). An Iterative Closest Point (ICP) with a probabilistic implementation is then used to register the point clouds, taking into account their uncertainties. The registration process is divided in two steps: (1) point-to-point association for coarse registration and (2) point-to-plane association for fine registration. The point clouds of the surfaces to be registered are sub-sampled in order to decrease both the computation time and also the potential of falling into local minima during the registration. In addition, a heuristic is used to decrease the complexity of the association step of the ICP from O(n2) to O(n). The performance of the SLAM framework is tested using two real world datasets: First, a 2.5D bathymetric dataset obtained with the usual down-looking multibeam sonar configuration, and second, a full 3D underwater dataset acquired with a multibeam sonar mounted on a pan and tilt unit This work was supported by the Spanish project DPI2014-57746-C3-3-R (ARCHROV) and two European Commission’s Seventh Framework Program projects: FP7-ICT-2011-7-288704 (MORPH) and FP7-INFRASTRUCTURES-2012-312762 (EUROFLEETS2)

We describe a model-based objects recognition system which is part of an image interpretation system intended to assist autonomous vehicles navigation. The system is intended to operate in man-made environments. Behavior-based navigation of autonomous vehicles involves the recognition of navigable areas and the potential obstacles. The recognition system integrates color, shape and texture information together with the location of the vanishing point. The recognition process starts from some prior scene knowledge, that is, a generic model of the expected scene and the potential objects. The recognition system constitutes an approach where different low-level vision techniques extract a multitude of image descriptors which are then analyzed using a rule-based reasoning system to interpret the image content. This system has been implemented using CEES, the C++ embedded expert system shell developed in the Systems Engineering and Automatic Control Laboratory (University of Girona) as a specific rule-based problem solving tool. It has been especially conceived for supporting cooperative expert systems, and uses the object oriented programming paradigm.

Research article compendium : peace and conflict research The University of Barcelona, in collaboration with different Spanish and foreign institutions and universities, hosted an international workshop on the ‘Sense and Scope of Autonomy in Emerging Military and Security Technologies’, which took place in February 2017. This academic event paved the way to the elaboration of the present report, which contains a compendium of innovative articles written by experts on scientific, legal, diplomatic and military matters. All papers reflect a broad spectrum of views on how ‘autonomy’ can be understood in emerging military and security technologies and what their legal, technical, political and societal impacts are by building on discussions unfolded along a variety of reflective approaches.

It is at least two decades since the conventional robotic manipulators have become a common manufacturing tool for different industries, from automotive to pharmaceutical. The advances in manipulators and sensors have given robots the opportunity to become useful for more and more applications. Engineers have taken advantage of the extra mobility of the advanced robots to make them work in constrained environments, ranging from limited joint motions for redundant manipulators to obstacles in the way of mobile (ground, marine, and aerial) robots [1]. However, the incorporation some of these abilities and capacities that are already being used in land, have not made their way to the sea domain. This Abstract describes the project consisting in the design, development and manufacture of a prototype manipulator arm for ROVs introducing innovative fabrication technologies. The work has been done collaboratively among ACSM Maritime Agency SL, CIMA Group and the University of Vigo.

Actualment el món on vivim és un món cada cop més automatitzat a causa del desenvolupament de les noves tecnologies. Cada vegada apareixen més robots automatitzats per tal de realitzar accions que necessiten molta precisió, el cas més concret és en la indústria. Tanmateix, també podem observar un augment considerable de petits robots per realitzar tasques diàries, com ara robots aspiradors autònoms. Fins i tot han aparegut en els darrers anys, diverses joguines automatitzades que simulaven animals de companyia, o petites avionetes amb control remot. Per aquest motiu, ha crescut l’interès per la introducció a la robòtica en l’educació. També cal destacar el creixement de diverses plataformes robòtiques de baix cost, que han permès desenvolupar el camp de la robòtica educativa. Els dos casos més concrets són Arduino, que és una placa electrònica de codi obert que permet controlar servomotors i/o sensors de forma molt senzilla, i LEGO Mindstorms, que és una plataforma centrada en la robòtica de la marca LEGO Today the world we live in is an increasingly automated world because of the development of new technologies. More and more automated robots are appearing to perform actions that require great precision, and the most specific case is in the industry. However, we can also see a considerable increase in small robots for daily tasks, such as freelance vacuum cleaners. Even in recent years, several automated toys that have been simulated for pets, or small drones with remote control, have appeared. For this reason, interest in introducing robotics in education has grown. Also noteworthy is the growth in various low-cost robotic platforms, which have allowed the field of educational robotics to develop. The two most specific cases are Arduino, which is an open source electronic board that allows servomotors and / or sensors to be controlled very simply, and LEGO Mindstorms, which is a platform focused on LEGO brand robotics

Marie Sklodowska-Curie grant agreement no. 750063 This paper presents a control architecture for an underwater vehicle-manipulator system (UVMS) to enable simultaneous tracking of end-effector configuration and contact force during floating-base manipulation. The main feature of the architecture is its combination of a task-priority (TP) kinematic control algorithm with a custom force control strategy, based on impedance (admittance) control. The TP algorithm used in the work includes recent treatment of equality and inequality tasks as well as original concepts to handle operation in singular configurations of the system. In the force control part the impedance concept is extended to allow for direct control over the value of exerted force and torque. Additional feed-forward signal is used to ensure stable contact. The performance of the control architecture is demonstrated by experiments in a test tank, with GIRONA500 I-AUV performing pipe inspection This work is part of a project titled ”Force/position control system to enable compliant manipulation from a floating IAUV”, which received funding from the European Union’s Horizon 2020 research and innovation programme, under the Marie Sklodowska-Curie grant agreement no. 750063