Search

El presente Trabajo de Fin de Grado se centra en la aplicación de técnicas de aprendizaje profundo (Deep Learning) a través del entrenamiento de Redes Neuronales Convolucionales para optimizar el diagnóstico del cáncer de pulmón en su etapa de detección y clasificación. El objetivo es crear un sistema de clasificación de imágenes de tejido pulmonar que permita una detección más precisa y temprana del cáncer, lo que a su vez contribuiría a mejorar la selección del tratamiento más adecuado. El proyecto se lleva a cabo en la plataforma Visual Studio Code utilizando el lenguaje Python junto a las librerías ofrecidas por la API de aprendizaje profundo Keras y una selección de la base de datos (LC25000) que cuenta con a 15000 imágenes de histologías de tejido pulmonar. Se entrenan tres modelos distintos de visión y se analizan sus resultados para ver cual presenta un mejor comportamiento en esta aplicación. El uso de este tipo de técnicas en el campo de la medicina está en constante crecimiento debido a su capacidad para aprender patrones complejos a partir de grandes conjuntos de datos. El éxito del proyecto proporcionará una herramienta útil y eficaz para el diagnóstico temprano de esta enfermedad pulmonar, lo que tendrá un impacto significativo en la salud y la calidad de vida de los pacientes., This thesis focuses on the application of deep learning techniques through the training of Convolutional Neural Networks to optimize the diagnosis of lung cancer in its detection and classification stage. The objective is to create a classification system for lung tissue images that will allow a more accurate and early detection of cancer, which in turn would contribute to improve the selection of the most appropriate treatment. The project is carried out on the Visual Studio Code platform using the Python language together with the libraries offered by the Keras deep learning API and a selection of the database (LC25000) which has 15000 lung tissue histology images. Three different vision models are trained, and their results are analyzed to see which one performs better in this application. The use of such techniques in the medical field is constantly growing due to their ability to learn complex patterns from large data sets. The success of the project will provide a useful and effective tool for the early diagnosis of this lung disease, which will have a significant impact on the health and quality of life of patients.

El objetivo de este Trabajo de fin de carrera es el de estudiar las ecuaciones que permiten realizar tanto el control como el modelado de un vehículo aéreo no tripulado tipo tilting quadrotor que contará con dos servomotores, es decir, un dron de 4 hélices donde los rotores serán inclinables dos a dos. Para ello, este documento se ha dividido en 3 partes fundamentales, referidas a cada uno de los pasos tomados en este estudio, que se han elaborado para obtener los resultados finales. En la primera de ellas, se realizará un estudio previo de las ecuaciones que modelarán tanto la dinámica como la cinemática del mismo. Además, se abordará la búsqueda de soluciones ante el problema de control que supone el incluir dos servomotores a la configuración. Posteriormente, se continuará modificando un simulador de vuelo según los resultados obtenidos en el apartado anterior. Por último, se hará una descripción de los puntos singulares que presenta el modelo dándoles, a su vez, un porqué. Para corroborar la veracidad y funcionalidad de todo lo anterior, se finalizará este trabajo llevando a cabo una serie de simulaciones que permitirán observar el comportamiento del dron frente a diferentes situaciones. A partir de los resultados, se elaborarán las conclusiones pertinentes., The aim of this final degree report is to study the equations that lead to the development of the control and modeling of a tilting quadrotor which has the ability to rotate its motors in pairs. To carry out this task, the document has been divided into three parts, related to each phases that that must be undertaken in order to reach the final objetive and obtain results. Firstly, a preliminary study of the equations that models the kinematics and dynamics of the system has been done. Additionally, this part includes a search for solutions that allows the calculations of the new variables introduced by the servomotors, the orientations of the motors. Secondly, the study continues with the modification of a flight simulator, which is adapted to incorporate the equations of the quadrotor. Finally, a description of the singularities presented by the model will be provided, along with a rationale for each. To verify the accuracy and functionality of the aforementioned work, this report will conclude with a series of simulations which allow making observations of the quadrotor’s behavior facing different situations. These results lead to make conclusions.

Este trabajo de fin de grado tiene como objetivo proponer un modelo analítico para estudiar la interacción física en vibración entre un UAV omnidireccional y un sistema flexible, específicamente un árbol. Surge a partir de los experimentos realizados por Antonio González Morgado durante su estancia de investigación en el ETH Zürich. El modelo analítico desarrollado en este trabajo fin de grado se basa en el método de Rayleigh, gracias al cual se puede aproximar un sistema continuo, como es el caso de un árbol, por un sistema de 1GdL (Grado de Libertad). Para ello, el árbol se divide en diferentes elementos como son el tronco, las ramas y los frutos. Posteriormente, utilizando el método de Rayleigh, se obtienen la masa y rigidez equivalente de los elementos que conforman el árbol. Posteriormente, se presentan diversas técnicas para inducir vibraciones a la frecuencia natural en un sistema mecánico. De todas las soluciones presentadas, se utilizó una estrategia de vibraciones autoinducidas (self-excited vibration en inglés) para los experimentos. Finalmente, el modelo analítico propuesto para el árbol es contrastado con los experimentos disponibles de la estancia de Antonio González Morgado en el ETH Zürich. Los resultados obtenidos muestran la concordancia entre el modelo teórico presentado y los datos experimentales, lo que indica que el enfoque utilizado para modelar el sistema es válido y confiable., The objective of this final degree project is to propose an analytical model to study the physical interaction in vibration between an omnidirectional UAV and a flexible system, specifically a tree. It derives from the experiments carried out by Antonio González Morgado during his research stay at the ETH Zürich. The analytical model developed in this final degree project is based on Rayleigh’s method, which allows to approximate a continuous system, such as a tree, by a 1DoF (Degree of Freedom) system. For this purpose, the tree is divided into different elements such as trunk, branches and fruits. Subsequently, using Rayleigh’s method, the equivalent mass and stiffness of the elements that form the tree are obtained. Subsequently, several techniques to induce vibrations at natural frequency in a mechanical system are presented. Of all the solutions presented, a self-excited vibration strategy was used for the experiments. Finally, the analytical model proposed for the tree is contrasted with the experiments available from the research stay of Antonio Gonzalez Morgado at ETH Zürich. The results obtained show the concordance between the theoretical model presented and the experimental data, which indicates that the approach used to model the system is valid and reliable.

En los últimos años se ha incrementado el uso de robots aéreos en las tareas de inspección. Las plantas industriales cuentan con cientos de metros de tuberías y deben revisarlas con elevada frecuencia en busca de posibles fallas o derrames. Estas inspecciones, previamente realizadas de forma manual, se realizan de forma más veloz y eficiente a través de UAVs. Además, se evita poner en riesgo la vida de seres humanos, dado que las tuberías se ubican comúnmente en zonas altas o poco accesibles. La solución planteada en el trabajo se basa en un UAV que deposita un crawler (o "carrito") en la tubería. Este dispositivo, equipado con sensores, realiza las comprobaciones necesarias y, posteriormente, es recogido por el dron mediante una cuerda. Este documento se centra en el despegue del UAV sobre un conducto. Para desarrollar la misión, esta se ha divido en dos fases, la primera de ellas estudia el comportamiento del UAV atado, y la segunda, el vuelo del sistema UAV+crawler. Para cada una de estas etapas se han estudiado las ecuaciones de la dinámica y el control. Posteriormente, se implementaron en Simulink dichas ecuaciones para poder simular el funcionamiento del sistema, obteniendo gráficas que aportan información relevante. Además de las simulaciones, se realizó un programa en Matlab que muestra el recorrido del quadrotor en un vídeo. En el primero de los simuladores se empleó la segunda Ley de Newton para obtener las ecuaciones de la dinámica y un sistema con feedback linearization para implementar el control, el cual permite transformar la dinámica de un sistema no lineal en una dinámica lineal para poder aplicar las estrategias de control lineal. Para el simulador del vuelo se utilizaron las ecuaciones de Euler-Lagrange en la dinámica y una estrategia de control conocida como Interconnection and Damping Assignment-Passivity Based Control (IDA-PBC), que sirve para estabilizar sistemas lineales con estructura de bucle cerrado a través del Hamiltoniano. Para ello, el, In the past few years, the use of aerial robots in inspection tasks has increased. Industrial plants have hundreds of meters of pipelines which must be checked often for possible spills. These inspections used to be performed manually, however, UAVs have proven to carry them out more quickly and efficiently. Moreover, drones prevent from risking human lifes, as these pipelines are commonly located at high altitudes or hazardous places. The solution studied in this project is based on a drone that deposits a crawler on the pipe. This device, equipped with sensors, performs the inspection. Afterwards, the crawler is picked up by the UAV thanks to a cable. The following document focuses on the take-off of the UAV over the pipeline. Thereby, to perform the task correctly, it has been divided in two phases, firstly, the tethered UAV phase, and secondly, the flight of the system composed of UAV+Payload. For each of these stages, the dynamics and control equations have been studied, afterwards, they have been implemented in Simulink, in order to check the correct operation of the system through simulations that allow to comprehend its behaviour and obtain graphics that provide useful information. Apart from the simulations, a Matlab code is also added to show the trajectory followed by the quadrotor during the flight through a generated video. For the first simulator, the dynamics equations were obtained through Newton’s Second Law, while the control strategy used a feedback linearization, which allows to transform the dynamics of a non-lineal system into a lineal system, so that we can apply the lineal control techniques. Furthermore, for the flight simulator, the dynamics were studied thanks to the lagrangian equations and the control strategy assumed this time was an Interconnection and Damping Assignment-Passivity Based Control (IDA-PBC), which is useful to stabilize lineal systems with a closed-loop structure using the Hamiltonian. IDA-PCB achieves this goal injecting

Con el desarrollo de la robótica aérea han aparecido nuevas plataformas de multirotores de actuación completa (fully-actuated en inglés), las cuales tienen la capacidad de desplazarse sin inclinar la plataforma. Este artículo presenta una comparación en cuanto a capacidades de movimiento entre un hexarotor de rotores coplanarios, configuración estándar, y un hexarotor de rotores inclinados, configuración fully-actuated. Para ello, se presenta el diseño, modelo y control de ambas configuraciones. Tras el montaje de las plataformas, se comparan con diferentes trayectorias, mediante simulaciones y experimentos. Asímismo, se muestran capacidades exclusivas de la plataforma fully-actuated, como la capacidad de mantenerse en hover con un ángulo de inclinación. Finalmente, se presenta la aplicación de la plataforma fully-actuated para inspección visual de techos de puentes. Vídeo del artículo: https://youtu.be/d95Qvz5hba4, With the development of aerial robotics, new multi-rotor platforms, known as fully-actuated, have appeared. These platforms have the ability to move without tilting the platform. This article presents a comparison in terms of motion capabilities between a coplanar hexarotor, standard configuration, and a tilted-propellers hexarotor, fully-actuated configuration. For this purpose, this paper presents the design, modelling and control of both configurations. Both platforms are compared with different trajectories through simulations and experiments. Also, unique capabilities of the fully-actuated platform, such as the ability to hover at a tilt angle, are shown. Finally, the use of the fully-actuated platform for visual inspection of bridge beams is included. Video of the paper: https://youtu.be/d95Qvz5hba4

Motivated by the convenience of improving the performance of long-reach aerial manipulators in the realization of maintenance tasks on high-voltage power lines, this paper proposes a constrained design optimization method for dual-arm aerial manipulators intended to reduce the weight while increasing the workspace of the robot. This configuration, in which the arms are separated from the aerial platform through a long reach link similar to a pendulum, improves safety in the interaction with human workers, reduces the electromagnetic interference of high voltage power lines on the electronics, as well as the aerodynamic downwash effect due to the propellers. However, the long-reach link introduces undesired vibrations on the manipulator due to its flexibility, so its length imposes a trade-off between the safety of operation as a positive side-effect and vibration as a negative one. Therefore, the cost function in the optimization problem also accounts for this factor, limiting the vibration to a fixed predefined value. A recent optimization approach is used here to minimize the cost function and solve the problem, verified by particle swarm optimization as a basis to confirm the correctness of the obtained data.

En este Trabajo Fin de Grado se ha estudiado la influencia de la inercia de un brazo manipulador en un multirrotor utilizando un control desacoplado, es decir, un control independiente para el multirrotor y otro para el manipulador; ya que este es el esquema de control más extendido en multirrotores comerciales. Con este fin, se ha realizado en primer lugar un simulador, con objeto de definir unos límites aproximados de máxima inercia del manipulador y máxima velocidad de movimiento de este, a partir de los cuales el multirrotor deja de ser controlable con un esquema de control desacoplado. Tras esto, se ha realizado una verificación experimental de los resultados obtenidos en simulación, utilizando para ello un multirrotor con un manipulador de inercia variable diseñado expresamente para estos experimentos. Para facilitar los experimentos, se ha desarrollado una nueva Estación de Control en Tierra (GCS, Ground Control Station). Por último, una vez obtenidos los resultados experimentales, se ha construido otro brazo manipulador con la inercia adecuada atendiendo a los resultados anteriores, diseñado para realizar teleoperación. Para mejorar la precisión de la teleoperación, se ha construido otro brazo robótico que, junto a la GCS desarrollada, permite controlar el manipulador acoplado al UAV mediante la imitación de los movimientos de este nuevo brazo, que es movido por un operario desde Tierra., In this End of Degree Project the influence of a manipulator arm on a multirotor has been estudied. For this purpose, a decoupled control scheme has been implemented, i.e. an independent controller for the multirotor an another one for the manipulador; as this is the most common control scheme in commercial UAVs. With this goal, a simulator has been developed to define approximately the maximum inertia limits of the manipulator and its maximum speed of movement. Going further than them will cause the lost of control of the multirotor using a decoupled control scheme. After that, the results of the simulations has been approved by doing some experiments, using a multirotor with a variable inertia manipulator. To make easier these experiments, a new Ground Control Station has been developed. Once the experimental results have been obtained, another manipulator arm has been built in order to do teleoperation tasks. To improve the accuracy of these tasks, another robotic arm has been built, not as large as the one on the UAV. With this smaller arm and the GRVC GCS, it is posible to control the multirotor arm by copying the pose of the other arm, which is moved directly by an operator.

Este trabajo propone una solución frente a un problema presente en los brazos robóticos flexibles: la deflexión. Se ha desarrollado un sistema de visión que utiliza varios métodos simultáneamente, buscando la mayor precisión y fiabilidad a la hora de medirla. Los métodos utilizados son: • la detección por umbralización del color: Se utilizará con el objetivo de encontrar un marker rojo en la muñeca del robot. • filtrado de fondo por profundidad: Aprovecha el conocimiento de las dimensiones del robot para su desempeño • CamShift: Algoritmo ampliamente utilizado en el tracking de objetos por su color. Se calculará la deflexión como la diferencia entre la posición cartesiana del marker, proporcionada por el sistema de visión, y los puntos obtenidos del modelo rígido, calculados del modelo cinemático del robot. Mediante dos experimentos distintos en un entorno controlado se ha comprobado la eficacia del sistema de visión en la precisión de las medidas obtenidas por la cámara y su cálculo de la deflexión., This Bachelor Thesis proposes a solution to a common problem in flexible robotic arms: the deflection. A vision system has been developed including several methods simultaneously, seeking the highest accuracy and reliability when measuring it. The methods used are: • detection by color thresholding: It will be used with the objective of finding a red marker on the wrist of the robot. • depth background filtering: It takes advantage of the knowledge of the robot’s dimensions for its performance. • CamShift: Algorithm widely used in object tracking by color. . The deflection will be calculated as the difference between the Cartesian position of the marker, provided by the vision system, and the points obtained from the stiff-joint model, obtained from the kinematic model of the robot. Two experiments in a controlled environment tested the effectiveness of the vision system in the accuracy of the measurements obtained by the camera and its deflection calculation.

This work proposes a novel proportional-derivative (PD)-type state-dependent Riccati equation (SDRE) approach with iterative learning control (ILC) augmentation. On the one hand, the PD-type control gains could adopt many useful available criteria and tools of conventional PD controllers. On the other hand, the SDRE adds nonlinear and optimality characteristics to the controller, i.e., increasing the stability margins. These advantages with the ILC correction part deliver a precise control law with the capability of error reduction by learning. The SDRE provides a symmetric-positive-definite distributed nonlinear suboptimal gain K(x) for the control input law u = −R−1(x)BT(x)K(x)x. The sub-blocks of the overall gain R−1(x)BT(x)K(x), are not necessarily symmetric positive definite. A new design is proposed to transform the optimal gain into two symmetric-positive-definite gains like PD-type controllers as u = − KSP(x)e-KSD(x)e. The new form allows us to analytically prove the stability of the proposed learning-based controller for mechanical systems; and presents guaranteed uniform boundedness in finite-time between learning loops. The symmetric PD-type controller is also developed for the state-dependent differential Riccati equation (SDDRE) to manipulate the final time. The SDDRE expresses a differential equation with a final boundary condition, which imposes a constraint on time that could be used for finite-time control. So, the availability of PD-type finite-time control is an asset for enhancing the conventional classical linear controllers with this tool. The learning rules benefit from the gradient descent method for both regulation and tracking cases. One of the advantages of this approach is a guaranteed-stability even from the first loop of learning. A mechanical manipulator, as an illustrative example, was simulated for both regulation and tracking problems. Successful experimental validation was done to show the capability of the system in practice by th

El objetivo fundamental de este trabajo de fin de grado es el estudio del efecto suelo en un hexarotor, configurado tanto con rotores inclinados como para rotores coplanarios. Para llevar a cabo este estudio, se comienza importando el modelo de Matlab-Simulink desarrollado en el trabajo de fin de grado de mi tutor, Don Antonio González Morgado, al cual se le aplican las modificaciones pertinentes para poder corregir el efecto suelo. Con este modelo se simulan diferentes situaciones en las que se observan las perturbaciones creadas por el efecto suelo, tanto para la configuración de rotores coplanarios como para la de rotores inclinados. Tras mostrar en estas simulaciones las perturbaciones en el control debidas al efecto suelo, se realiza un estudio bibliográfico con las soluciones desarrolladas para hacer frente al vuelo bajo efecto suelo. Tras analizar y explicar las ventajas e inconvenientes de cada una de ellas, en este Trabajo Fin de Grado se propone una nueva solución: un controlador de fuerza de bajo nivel en cada rotor. Este método se implementa en el anterior simulador de Maltab-Simulink y, mediante diferentes simulaciones, se demuestra la mejora en el control con el método propuesto., The main objective of this final degree project is the study of the ground effect in a hexarotor, using both configuration: tilted rotors configuration and coplanar rotors configuration. To carry out this study, the first step was to import the Matlab-Simulink model used on the final degree project of my tutor, Mr. Antonio González Morgado, to which some modifications were applied in order to correct the ground effect. With this model, different situations are simulated in which the disturbances created by the ground effect are observed, both for the configuration of coplanar rotors and for the configuration of tilted rotors. After showing in these simulations the control disturbances due to ground effect, a bibliographic study is made with the solutions developed to cope with flight under ground effect. After analyzing and explaining the advantages and disadvantages of each of them, a new solution is proposed in this Final Degree Project: a low level force controller in each rotor. This method is implemented in the previous Maltab-Simulink simulator and, through different simulations, the improvement in control with the proposed method is demonstrated.

Tren de aterrizaje, acoplable a una superficie inferior de un dron (3), para posicionar sensores sobre una tubería para inspección, que comprende al menos dos mecanismos de pinza (1), acoplables a la superficie inferior del dron (3), cada uno con una estructura de acoplamiento (4) conectada con un cuerpo central (5); dos brazos (6) laterales, que pueden pivotar alrededor de un extremo interior unido al cuerpo central (5), y que comprenden ruedas omnidireccionales (7), acopladas a su cara interna; un primer actuador (9), que mueve los brazos (6); y un mecanismo de rueda tractora (2) orientable, acoplado a la zona central de la superficie inferior del dron (3), mediante una plataforma fija (10), alrededor de la que gira una plataforma móvil (11), mediante un segundo actuador (12); con una rueda tractora (13), unida a la plataforma móvil (11) mediante una primera suspensión (14) articulada, accionada por un tercer actuador (15).

Con el desarrollo de la robótica aérea han aparecido nuevas plataformas de multirotores de actuación completa (fully-actuated en inglés), las cuales tienen la capacidad de desplazarse sin inclinar la plataforma. Este artículo presenta una comparación en cuanto a capacidades de movimiento entre un hexarotor de rotores coplanarios, configuracióon estándar, y un hexarotor de rotores inclinados, configuración fully-actuated. Para ello, se presenta el diseño, modelo y control de ambas configuraciones. Tras el montaje de las plataformas, se comparan con diferentes trayectorias. Así mismo, se muestran capacidades exclusivas de la plataforma fully-actuated, como la capacidad de mantenerse en hover con un ángulo de inclinación.

This paper presents an aerial manipulation robot consisting of a hexa-rotor equipped with a 2-DOF (degree of freedom) Cartesian base (XY–axes) that supports a 1-DOF compliant joint arm that integrates a gripper and an elastic linear force sensor. The proposed kinematic configuration improves the positioning accuracy of the end effector with respect to robotic arms with revolute joints, where each coordinate of the Cartesian position depends on all the joint angles. The Cartesian base reduces the inertia of the manipulator and the energy consumption since it does not need to lift its own weight. Consequently, the required torque is lower and, thus, the weight of the actuators. The linear and angular deflection sensors of the arm allow the estimation, monitoring and control of the interaction wrenches exerted in two axes (XZ) at the end effector. The kinematic and dynamic models are derived and compared with respect to a revolute-joint arm, proposing a force-position control scheme for the aerial robot. A battery counterweight mechanism is also incorporated in the X–axis linear guide to partially compensate for the motion of the manipulator. Experimental results indoors and outdoors show the performance of the robot, including object grasping and retrieval, contact force control, and force monitoring in grabbing situations.

Hoy en día, el sector de los UAVs (vehículos aéreos no tripulados) está creciendo continuamente. Cada vez se encuentran más aplicaciones para este tipo de aeronaves, como por ejemplo vigilancia de fronteras, grabación de eventos, transporte urgente de órganos para operaciones de trasplante, etc. Es más, se diseñan drones cuya configuración y geometría están optimizadas para desempeñar una función específica. En este contexto surgen los multirrotores quadrotor como uno de los UAV más populares y más habituales para la resolución de tareas y misiones propuestas a los vehículos aéreos no tripulados. El objetivo de este proyecto es definir el comportamiento y posteriormente realizar el control de un multirrotor quadrotor destinado a la inspección de túneles. De esta forma, en primer lugar se modela la dinámica del sistema y se obtienen las ecuaciones que gobiernan el movimiento del UAV para posteriormente realizar su control y comprobar su funcionamiento por medio de simulaciones., These days, the UAV (Unmanned Aerial Vehicle) sector is continually growing. There are more and more applications for this type of aircraft, such as border surveillance, event recording, urgent transport of organs for transplantation operations, etc. What’s more, drones are designed whose configuration and geometry are optimized to perform a specific function. In this context, quadrotor multirotors emerge as one of the most popular and common UAVs for solving tasks and missions proposed to unmanned aerial vehicles. The objective of this project is to define the behavior and subsequently carry out the control of a quadrotor multirotor for tunnel inspection. In this way, in the first place the dynamics of the system is modeled and the equations that govern the movement of the UAV are obtained to later carry out its control and check its operation by means of simulations

El objetivo principal que busca este trabajo de fin de grado consiste en el estudio del comportamiento en simulación de un sistema multirrotor de empuje bidireccional. Para llevar a cabo el estudio es necesario la construcción de un entorno de simulación para realizar los distintos ensayos planteados, justificándose con ello la necesidad de construir dicho entorno de simulación. Este proyecto recoge todo el proceso de construcción del modelo, desde el desarrollo de las ecuaciones que modelan el sistema, hasta su implementación en el simulador. Así mismo, se recogen el proceso y trabajo de las distintas operaciones que debe llevar a cabo el sistema. Una vez queda programado, se realizarán diversas simulaciones para observar la respuesta del sistema en distintas situaciones donde debe llevar a cabo las operaciones especificadas, realizando un análisis detallado del mismo y obteniendo conclusiones., The main objective of this final degree work is the study of the simulation behaviour of a bidirectional thrust multirotor system. In order to carry out the study it is necessary to build a simulation environment to perform the different tests proposed, thus justifying the need to build such simulation environment. This project includes the whole process of building the model, from the development of the equations that model the system to its implementation in the simulator. Likewise, the process and work of the different operations to be carried out by the system are included. Once it is programmed, several simulations will be carried out to observe the response of the system in different situations where it must carry out the specified operations, making a detailed analysis of it and obtaining conclusions.

This paper presents a localization system for Unmanned Aerial Vehicles (UAVs) especially designed to be used in infrastructure inspection, where the UAVs have to fly in challenging conditions, such as relatively high altitude (e.g., 15 m), eventually with poor or absent GNSS (Global Navigation Satellite System) signal reception, or the need for a BVLOS (Beyond Visual Line of Sight) operation in some periods. In addition, these infrastructure inspection applications impose the following requirements for the localization system: defect traceability, accuracy, reliability, and fault tolerance. Our system proposes a lightweight solution combining multiple stereo cameras with a robotic total station to comply with these requirements, providing full-state estimation (i.e., position, orientation, and linear and angular velocities) in a fixed and time-persistent reference frame. Moreover, the system can align and fuse all sensor measurements in real-time at high frequency. We have integrated this localization system in our aerial platform, and we have tested its performance for inspection in a real-world viaduct scenario, where the UAV has to operate with poor or absent GNSS signal at high altitude.

This paper investigates the inspection-of-pipe topic in a new framework, by rotation around a pipe, peculiar to industrial sites and refineries. The evolution of the ultimate system requires prototype design and preliminary tests. A new benchmark has been designed and built to mimic the rotation around a pipe, with the main purpose of assessing the different types of rotors and control systems. The benchmark control system presents a mechatronics package including mechanical design and machining, electronics and motor drive, motor-blade installation, computer programming, and control implementation. The benchmark is also modular, working with two modes of one- and two-degree-of-freedom (DoF), easily interchangeable. To cover a full rotation, conventional fixed-pitch drones fail to provide negative thrusts; nonetheless, variable-pitch (VP) rotor quad- copters can produce that in both directions. A closed-loop nonlinear optimal method is chosen as a controller, so- called, “the state-dependent Riccati equation (SDRE)” approach. Optimal control policies are challenging for experimentation though it has been successfully done in this report. The advantage of the VP is also illustrated in a rotation plus radial motion in comparison with fixed-pitch rotors while a wind gust disturbs the inspection task. The proposed VP system compensated the disturbance while the fixed pitch was pushed away by the wind gust. The solution methods to the SDRE were mixed, a closed-form exact solution for the one-DoF system, and a numerical one for the two-DoF. Solving the Riccati online in each time step is a critical issue that was effectively solved by the implementation approach, through online communication with MATLAB software. Both simula- tions and experiments have been performed along with a discussion to prove the application of VP systems in rotary-motion pipe inspection

En este Trabajo Fin de Grado se lleva a cabo el desarrollo de los algoritmos de control del autopiloto para un UAV tipo quadrotor para la inspección de túneles. Debido a diversos motivos que se expondrán a lo largo del documento, este UAV estará cordado a una plataforma o robot terrestre, siendo necesario un análisis de la coordinación entre ambos. Tras realizar una revisión del estado actual de este tipo de tecnologías, se ha diseñado un controlador para el UAV y se ha analizado su validez a través de Simulink, mediante la simulación de los códigos implementados en MATLAB. Durante el trabajo se hace uso también de otros softwares como son CATIA V5 para obtener las características dinámicas del UAV y AutoCAD para el diseño de las trayectorias. Adicionalmente, se ha incluido una metodología basada en la aplicación de un campo artificial de fuerzas potenciales para evitar la colisión del UAV con posibles obstáculos que puedan existir dentro de los túneles. Finalizado el análisis se expondrán distintos ejemplos de aplicación del diseño realizado, con los cuales se demuestra el cumplimiento de los objetivos del proyecto y el perfecto funcionamiento del controlador en todo tipo de túneles y de trayectorias consideradas., In this Final Degree Project, the development of the control algorithms of the autopilot for a quadrotor UAV for tunnel inspection is carried out. Due to several reasons that will be exposed throughout the document, this UAV will be tethered to a platform or terrestrial robot, being necessary an analysis of the coordination between both. After conducting a review of the current state of this type of technology, a controller for the UAV has been designed and its validity has been analyzed through Simulink, by simulating the codes implemented in MATLAB. During the work, it is also made use of other softwares such as CATIA V5 to obtain the dynamic characteristics of the UAV and AUTOCAD for the design of the flight path. Additionally, a methodology based on the application of an artificial field of potential forces has been included to avoid the collision of the UAV with possible obstacles that may exist within the tunnels. After the analysis, different application examples of the carried out design will be presented, with which it is demonstrated the fulfillment of the project objectives and the perfect operation of the controller in all types of tunnels and flight path considered.

Este trabajo se centra en el problema completo de la estimación del estado en vehículos aéreos no tripulados (UAVs). Para ello, se presentan conceptos básicos necesarios como los sistemas de representación de la actitud y los sistemas de referencia comunes en vehículos aéreos. Además, también se describen herramientas esenciales como el álgebra de cuaterniones, el filtro de Kalman y su versión extendida para sistemas no lineales. Después de hacer un análisis exhaustivo de los sensores disponibles para la estimación, cómo utilizarlos, y cómo calibrarlos, se presentan y describen los algoritmos, la implementación y los resultados obtenidos al resolver los problemas de estimación de la actitud y la posición. En este último, se utiliza una combinación de sensores y algoritmos tal, que resuelve uno de los problemas más comunes en los sistemas de localización actuales. Con la solución propuesta se abre la puerta a la repetitividad operacional que, en aplicaciones como la inspección aérea, son necesarias para, por ejemplo, evaluar y monitorizar defectos concretos de una infraestructura a lo largo del tiempo. Finalmente, se muestra el resultado de un trabajo de integración en el que las técnicas de estimación presentadas se integran con trabajos de diseño y control para resolver un caso de uso específico de inspección por contacto en una infraestructura., T his master’s thesis is focused on the entire problem of state estimation in unmanned aerial vehicles (UAVs). For this, necessary basic concepts such as attitude representation systems and common reference systems in aerial vehicles are presented. In addition, essential tools such as quaternion algebra, the Kalman filter and its extended version for non-linear systems are also described. After doing a thorough analysis of the sensors available for estimation, how to use and calibrate them, the algorithms, implementation, and the obtained results when solving attitude and position estimation problems are presented and described. In the latter, such a combination of sensors and algorithms is used to solve one of the most common problems in current location systems. The proposed solution opens the door to operational repeatability that, in applications such as aerial inspection, are necessary to, for example, evaluate and monitor specific defects in an infrastructure over time. Finally, the result of an integration work is shown in which the estimation techniques presented are integrated with design and control work to solve a specific use case of contact inspection in an infrastructure

The interest of the research community in Unmanned Aerial Vehicles (UAVs) has signi cantly grown in the last decade. Actually, this topic is becoming specially attractive for industrial companies due to its varied application range. Beyond their current commercial applications, where drones are used to carry out passive tasks, such as lming, surveillance, visual inspection, monitoring or remote sensing, the research community is going further using these drones as aerial robots which can physically interact with the environment. In those interactions, the aerial robot works quite close to several obstacles or objects, and thus, the aerodynamic characterisation of the phenomena that arise when ying close to those obstacles becomes relevant to ensure that the aerial manipulation operation can be performed in a safe and e ective way. This thesis takes part in this speci c problem that arises in aerial manipulators ying close to obstacles, through the development of new models and adapted control techniques to ensure safety and e ectiveness during operation in real scenarios. Along this Thesis, the author proposes the study of the generalisation of the classical method based on linearised potential aerodynamics and also through experimental results. The Thesis also treats two di erent cases of use in which the aerodynamic e ect has been taken into account from the control point of view. Last, it makes a twist to the modelled aerodynamic e ect going beyond classical models, studying how di erent aerodynamic e ects arise when the robot approaches di erent obstacles.

This paper considers the problem of performing bimanual aerial manipulation tasks in grabbing conditions, with one of the arms grabbed to a fixed point (grabbing arm) while the other conducts the task (operation arm). The goal was to evaluate the positioning accuracy of the aerial platform and the end effector when the grabbing arm is used as position sensor, as well as to analyze the behavior of the robot during the aerial physical interaction on flight. The paper proposed a control scheme that exploits the information provided by the joint sensors of the grabbing arm for estimating the relative position of the aerial platform w.r.t. (with respect to) the grabbing point. A deflection-based Cartesian impedance control was designed for the compliant arm, allowing the generation of forces that help the aerial platform to maintain the reference position when it is disturbed due to external forces. The proposed methods were validated in an indoor testbed with a lightweight and compliant dual arm aerial manipulation robot.

The ability of aerial manipulation robots to reach and operate in high altitude workspaces may result of interest in a wide variety of applications and scenarios that nowadays cannot be accessed easily by human operators. Consider for example the installation of sensors in polluted areas, the insulation of leaks in pipe structures, or the corrosion repair in power lines and wind turbines. This paper describes the application of a human-like dual arm aerial manipulator for the inspection of pipe structures, typical of chemical plants, involving the installation and retrieval of sensor devices. The goal is to reduce the time, cost, and risk with respect to conventional solutions conducted by human workers. Two con gurations of the aerial robot are considered and compared: the standard, in which the arms are installed at the base of the multirotor, and the long reach con guration in passive pendulum, which extends the effective workspace of the manipulator and increases safety during the operation on ight. The kinematic and dynamic models of both con gurations are derived, proposing a uni ed notation for the equations of motion, and a force/position control scheme that relies on the servo controller and the mechanical joint compliance. The paper also describes a simulation framework used for validating the execution of the aerial manipulation task before the realization of the real experiments, which contributes to reducing the probability of failure. The potential application of the standard and long reach con gurations is evaluated in two sensor installation tasks carried out in an indoor testbed

This paper presents the design, modeling and control of a fully actuated aerial robot for infrastructure contact inspection as well as its localization system. Health assessment of transport infrastructure involves measurements with sensors in contact with the bridge and tunnel surfaces and the installation of monitoring sensing devices at specific points. The design of the aerial robot presented in the paper includes a 3DoF lightweight arm with a sensorized passive joint which can measure the contact force to regulate the force applied with the sensor on the structure. The aerial platform has been designed with tilted propellers to be fully actuated, achieving independent attitude and position control. It also mounts a “docking gear” to establish full contact with the infrastructure during the inspection, minimizing the measurement errors derived from the motion of the aerial platform and allowing full contact with the surface regardless of its condition (smooth, rough, ...). The localization system of the aerial robot uses multi-sensor fusion of the measurements of a topographic laser sensor on the ground and a tracking camera and inertial sensors on-board the aerial robot, to be able to fly under the bridge deck or close to the bridge pillars where GNSS satellite signals are not available. The paper also presents the modeling and control of the aerial robot. Validation experiments of the localization system and the control system, and with the aerial robot inspecting a real bridge are also included.

This paper presents the design and development of a winged aerial robot with bimanual manipulation capabilities, motivated by the current limitations of aerial manipulators based on multirotor platforms in terms of safety and range/endurance. Since the combination of gliding and flapping wings is more energy e cient in forward flight, we propose a new morphology that exploits this feature and allows the realization of dexterous manipulation tasks once the aerial robot has landed or perched. The paper describes the design, development, and aerodynamic analysis of this winged aerial manipulation robot (WAMR), consisting of a small-scale dual arm used for manipulating and as a morphing wing. The arms, fuselage, and tail are covered by a nylon cloth that acts as a cap, similar to a kite. The three joints of the arms (shoulder yaw and pitch, elbow pitch) can be used to control the surface area and orientation and thus the aerodynamic wrenches induced over the cloth. The proposed concept design is extended to a flapping-wing aerial robot built with smart servo actuators and a similar frame structure, allowing the generation of di erent flapping patterns exploiting the embedded servo controller. Experimental and simulation results carried out with these two prototypes evaluate the manipulation capability and the possibility of gliding and flying.

This paper presents a novel method for motion planning of aerial long-reach manipulators that considers the aerodynamic effects generated by close surfaces in the trajectory generation process. The aerial manipulation system consists of a multirotor equipped with a robotic long-reach arm that enables multidirectional inspection and also increases considerably the safety distance between the rotors and the inspected elements. Since these systems operate in the proximity of elements that can modify significantly the rotors' airflow, the inclusion of Aerodynamics Awareness within the motion planning process is required to ensure robust obstacle avoidance. To this end, a proper characterisation of the aerodynamic effects based on both theoretical and experimental considerations has been derived. This characterisation is taken into account in the trajectory generation process to discard states whose associated aerodynamic phenomena are not well compensated by the system controller and to explore alternatives that lead to the most efficient trajectories within the area of safe operation. Moreover, the motion planner also stands out for three other relevant features: the joint consideration of the multirotor and the robotic long-reach arm, the generation of efficient trajectories in terms of energy consumption, and the Dynamics Awareness of the strong coupling between the aerial platform and the robotic arm. The resulting motion planner has been successfully tested in a simulated environment that faithfully reflects an application scenario strongly affected by aerodynamic effects: the inspection of bridges to find potential cracks in the surface of pillars.

This paper considers the inspection by contact of long arrays of pipe structures in hard-to-reach places, typical of chemical plants or oil and gas industries, presenting the design of a hybrid rolling-aerial platform capable of landing and moving along the pipes without wasting energy in the propellers during the inspection. The presented robot overcomes the limitation in terms of operation time and positioning accuracy in the application of fiying robots to industrial inspection and maintenance tasks. The robot consists of a hexa-rotor platform integrating a rolling base with velocity and direction control, and a 5-DOF (degree of freedom) robotic arm supported by a 1-DOF linear guide system that facilitates the deployment of the arm in the array of pipes to inspect their contour once the platform has landed. Given a set of points to be inspected in different arrays of pipes, the path of the multirotor and the rolling platform is planned with a hybrid RRT(Rapidly-exploring Random Tree) based algorithm that minimizes the energy consumption. The performance of the system is evaluated in an illustrative outdoor scenario with two arrays of pipes, using a laser tracking system to measure the position of the robot from the ground control station.

The aerial manipulation is a research field which proposes the integration of robotic manipulators in aerial platforms, typically multirotors – widely known as “drones” – or autonomous helicopters. The development of this technology is motivated by the convenience to reduce the time, cost and risk associated to the execution of certain operations or tasks in high altitude areas or difficult access workspaces. Some illustrative application examples are the detection and insulation of leaks in pipe structures in chemical plants, repairing the corrosion in the blades of wind turbines, the maintenance of power lines, or the installation and retrieval of sensor devices in polluted areas. Although nowadays it is possible to find a wide variety of commercial multirotor platforms with payloads from a few gramps up to several kilograms, and flight times around thirty minutes, the development of an aerial manipulator is still a technological challenge due to the strong requirements relative to the design of the manipulator in terms of very low weight, low inertia, dexterity, mechanical robustness and control. The main contribution of this thesis is the design, development and experimental validation of several prototypes of lightweight (<2 kg) and compliant manipulators to be integrated in multirotor platforms, including human-size dual arm systems, compliant joint arms equipped with human-like finger modules for grasping, and long reach aerial manipulators. Since it is expected that the aerial manipulator is capable to execute inspection and maintenance tasks in a similar way a human operator would do, this thesis proposes a bioinspired design approach, trying to replicate the human arm in terms of size, kinematics, mass distribution, and compliance. This last feature is actually one of the key concepts developed and exploited in this work. Introducing a flexible element such as springs or elastomers between the servos and the links extends the capabilities of the manipulato, La manipulación aérea es un campo de investigación que propone la integración de manipuladores robóticos in plataformas aéreas, típicamente multirotores – comúnmente conocidos como “drones” – o helicópteros autónomos. El desarrollo de esta tecnología está motivada por la conveniencia de reducir el tiempo, coste y riesgo asociado a la ejecución de ciertas operaciones o tareas en áreas de gran altura o espacios de trabajo de difícil acceso. Algunos ejemplos ilustrativos de aplicaciones son la detección y aislamiento de fugas en estructura de tuberías en plantas químicas, la reparación de la corrosión en las palas de aerogeneradores, el mantenimiento de líneas eléctricas, o la instalación y recuperación de sensores en zonas contaminadas. Aunque hoy en día es posible encontrar una amplia variedad de plataformas multirotor comerciales con cargas de pago desde unos pocos gramos hasta varios kilogramos, y tiempo de vuelo entorno a treinta minutos, el desarrollo de los manipuladores aéreos es todavía un desafío tecnológico debido a los exigentes requisitos relativos al diseño del manipulador en términos de muy bajo peso, baja inercia, destreza, robustez mecánica y control. La contribución principal de esta tesis es el diseño, desarrollo y validación experimental de varios prototipos de manipuladores de bajo peso (<2 kg) con capacidad de acomodación (“compliant”) para su integración en plataformas aéreas multirotor, incluyendo sistemas bi-brazo de tamaño humano, brazos robóticos de articulaciones flexibles con dedos antropomórficos para agarre, y manipuladores aéreos de largo alcance. Puesto que se prevé que el manipulador aéreo sea capaz de ejecutar tareas de inspección y mantenimiento de forma similar a como lo haría un operador humano, esta tesis propone un enfoque de diseño bio-inspirado, tratando de replicar el brazo humano en cuanto a tamaño, cinemática, distribución de masas y flexibilidad. Esta característica es de hecho uno de los conceptos clave desarrollados y uti

This paper presents a robotic system using Unmanned Aerial Vehicles (UAVs) for bridge-inspection tasks that require physical contact between the aerial platform and the bridge surfaces, such as beam-deflection analysis or measuring crack depth with an ultrasonic sensor. The proposed system takes advantage of the aerodynamic ceiling effect that arises when the multirotor gets close to the bridge surface. Moreover, this paper describes how a UAV can be used as a sensor that is able to fly and touch the bridge to take measurements during an inspection by contact. A practical application of the system involving the measurement of a bridge’s beam deflection using a laser tracking station is also presented. In order to validate our system, experiments on two different bridges involving the measurement of the deflection of their beams are shown.

Hace unos años, dentro de la robótica aérea, surgió la manipulación aérea como campo de investigación. Desde su nacimiento, su impacto ha ido incrementándose poco a poco debido, sobretodo, al gran número de aplicaciones que podrían llevarse a cabo con este tipo de sistemas. Un manipulador aéreo puede definirse como una plataforma aérea la cual ha sido equipada con uno o varios brazos robóticos. Este nuevo concepto ha abierto un mundo de posibilidades para este tipo de robots aéreos. Además, gracias a la posibilidad de este tipo de robots aéreos de interactuar con su entorno, podrían llevar a cabo inspecciones de estructuras civiles o incluso, tareas de ensamblaje de estructuras y todo ello, por supuesto, de forma autónoma. Esta tesis se centra en el estudio e implementación de sistemas de manipulación aérea y, en particular, en el diseño de estrategias de control para la plataforma aérea. Este estudio comienza con el cáculo de las ecuaciones que representan la dinámica del sistema, y que nos permite analizar su comportamiento y la influencia del movimiento de los brazos robóticos en la estabilidad de la plataforma.El análisis de estas ecuaciones nos permite diseñar de esquemas de control tales como los basados en Backstepping. Pero el objetivo de esta tesis no es solo el diseño sino también la implementación de estas técnicas de control en sistemas de manipulación aérea reales y con capacidad de llevar a cabo tareas de manipulación en escenarios al aire libre. La principales contribuciones de esta tesis son el cálculo de los modelos dinámicos para cada uno de los tipos de manipuladores aéreos estudiados he implementados durante el desarrollo de la tesis. Además del uso de estas modelos para la diseño de una estrategia de control adaptable a cada una de las plataformas. También se ha diseñado un mecanismo “compliant” que ha sido integrado en un manipulador parallevar a cabo tareas de inspección estructuras por contacto, además de un control de fuerza-posición. Cada m

El objetivo principal de este Trabajo de Fin de Grado es el estudio de los efectos aerodinámicos que se producen por el vuelo de rotores en el entorno de obstáculos. Además de eso, busca el desarrollo de un medio que permita la generación de un mapeado de los mismos para entornos definidos. Para esto, deberá hacerse uso de los estudios previos existentes sobre la caracterización de los efectos aerodinámicos mencionados. De igual forma, será necesaria la realización de ensayos experimentales, justificados por la inexistencia de resultados previos para algunos casos, completando de esta forma el conocimiento del comportamiento de los rotores en las proximidades de los obstáculos. Finalmente, el proyecto concluirá con el desarrollo de un programa encargado de implementar los datos recopilados. Permitiendo la obtención de los mapas de perturbaciones aerodinámicas con las condiciones del entorno definidas., The main objective of this Final Degree Project is the study of the aerodynamic effects which take place in the event of a rotor flying close to different obstacles. Moreover, it looks for the development of a mean which allows the generation of a map of them for known environments. In order to achieve that, it will be necessary making use of existing previous results related to the characterization of the aerodynamic effects mentioned above. Likewise, there will be need for carrying experiments, justified by the lack of previous results when it comes to certain situations, closing up the knowledge about the behaviour of a rotor in the surroundings of an obstacle. To conclude, the project will head to the development of a program which will implement all the gathered data. Allowing the user to obtain the aerodynamic perturbations map for the defined environment conditions.

El objetivo principal de este trabajo fin de grado es el estudio de la configuración de rotores inclinados en un hexarotor, así como su comparación con la configuración convencional de rotores coplanarios. Para llevar a cabo este estudio, se comienza con la obtención del modelo del sistema, el cual es necesario para realizar simulaciones. En este punto se comienza a observar las primeras diferencias con la configuración de rotores coplanarios. Obtenido el modelo del sistema, se llevan a cabo diferentes simulaciones, en las que se prueban diferentes controladores y se comparan entre sí las respuestas obtenidas. Así mismo, se compara la respuesta del hexarotor con rotores coplanarios. A continuación, se realiza un breve estudio del efecto suelo, así como los diferentes modelos que se puede usar para tenerlo en cuenta. Elegido un modelo del efecto suelo, se realizan simulaciones teniendo en cuenta el efecto suelo. En la última parte del proyecto, se lleva a cabo el diseño y montaje del hexarotor de hélices coplanarias. Para ello, se eligen los componentes y se lleva a cabo el diseño de las piezas necesarias para rotar los rotores., The main objective of this End-of-Degree Project is the study of a hexarotor UAV with tilted propellers, as well as its comparison with its standard configuration. To carry out the study, we start obtaining the system model, which is necessary to simulate the system. At this point, we can observe the first differences with the standard configuration. Oncewe have obtained themodel of the system, we performdifferent simulations, in which different controllers are tested and the obtained responses are compared. Moreover, we compare the response of the standard configuration with the tilted propellers configuration. Then, we carry out a brief study of ground effect and the different models we can use to consider it. Once we have chosen a ground effect model, we perform simulations where we consider the ground effect. Finally,we design and assemble the hexarotorUAVwith tilted propellers. To do this,we choose the components and we design the necessary piece for tilting the rotors.

This paper considers the problem of fault detection and identification (FDI) in applications carried out by a group of unmanned aerial vehicles (UAVs) with visual cameras. In many cases, the UAVs have cameras mounted onboard for other applications, and these cameras can be used as bearing-only sensors to estimate the relative orientation of another UAV. The idea is to exploit the redundant information provided by these sensors onboard each of the UAVs to increase safety and reliability, detecting faults on UAV internal sensors that cannot be detected by the UAVs themselves. Fault detection is based on the generation of residuals which compare the expected position of a UAV, considered as target, with the measurements taken by one or more UAVs acting as observers that are tracking the target UAV with their cameras. Depending on the available number of observers and the way they are used, a set of strategies and policies for fault detection are defined. When the target UAV is being visually tracked by two or more observers, it is possible to obtain an estimation of its 3D position that could replace damaged sensors. Accuracy and reliability of this vision-based cooperative virtual sensor (CVS) have been evaluated experimentally in a multivehicle indoor testbed with quadrotors, injecting faults on data to validate the proposed fault detection methods.

This paper presents an anthropomorphic, compliant and lightweight dual arm manipulator designed and developed for aerial manipulation applications with multi-rotor platforms. Each arm provides four degrees of freedom in a human-like kinematic configuration for end effector positioning: shoulder pitch, roll and yaw, and elbow pitch. The dual arm, weighting 1.3 kg in total, employs smart servo actuators and a customized and carefully designed aluminum frame structure manufactured by laser cut. The proposed design reduces the manufacturing cost as no computer numerical control machined part is used. Mechanical joint compliance is provided in all the joints, introducing a compact spring-lever transmission mechanism between the servo shaft and the links, integrating a potentiometer for measuring the deflection of the joints. The servo actuators are partially or fully isolated against impacts and overloads thanks to the ange bearings attached to the frame structure that support the rotation of the links and the deflection of the joints. This simple mechanism increases the robustness of the arms and safety in the physical interactions between the aerial robot and the environment. The developed manipulator has been validated through different experiments in fixed base test-bench and in outdoor flight tests.

This paper presents the development and experimental validation of a low weight and inertia, human-size and highly dexterous dual arm system designed for aerial manipulation with multirotor platform. The arms, weighting 1.8 kg in total and with a maximum lift load per arm around 0.75 kg, provide five degrees of freedom (DOF) for end-effector positioning and wrist orientation. A customized aluminum frame structure supports the servo actuators, placing most part of the mass close to the shoulder structure in order to reduce the inertia. A double flange bearing mechanism in side-by-side configuration isolates the servos from impacts and radial/axial overloads, increasing robustness. This is important to prevent that the arms are damaged during physical interactions with the environment, as they should support the kinetic energy of the whole platform. The motivation in the development of a dual arm aerial manipulator is extending the range of applications and tasks that can be performed with respect to the single arm case, like grasping large objects or assembling. The paper covers the kinematic and dynamic modeling of the aerial robot, proposing a control scheme that deals with the technological limitations of the smart servo actuators. The performance of the arms and the interactions with the aerial platform are evaluated in test bench experiments. The proposed dual arm design is validated through outdoor flight tests with two commercial hexarotor platforms equipped with standard industrial autopilots.

T his article summarizes new aerial robotic manipulation technologies and methods—aerial robotic manipulators with dual arms and multidirectional thrusters—developed in the AEROARMS project for outdoor industrial inspection and maintenance (I&M). Our report deals with -- the control systems, including the control of the interaction forces and the compliance -- the teleoperation, which uses passivity to tackle the tradeoff between stability and performance -- the perception methods for localization, mapping, and inspection -- the planning methods, including a new control-aware approach for aerial manipulation. Finally, we describe a novel industrial platform with multidirectional thrusters and a new arm design to increase the robustness in industrial contact inspections. In addition, the lessons learned in applying the platform to outdoor aerial manipulation for I&M are pointed out.

© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This article summarizes new aerial robotic manipulation technologies and methods-aerial robotic manipulators with dual arms and multidirectional thrusters-developed in the AEROARMS project for outdoor industrial inspection and maintenance (I&M), Peer Reviewed, Postprint (author's final draft)

This paper analyzes the ground effect in multirotors, that is, the change in the thrust generated by the rotors when flying close to the ground due to the interaction of the rotor airflow with the ground surface. This effect is well known in single-rotor helicopters but has been assumed erroneously to be similar for multirotors in many cases in the literature. In this paper, the ground effect for multirotors is characterized with experimental tests in several cases and the partial ground effect, a situation in which one or some of the rotors of the multirotor (but not all) are under the ground effect, is also characterized. The influence of the different cases of ground effect in multirotor control is then studied with several control approaches in simulation and validated with experiments in a test bench and with outdoor flights.

Este artículo presenta el primer robot manipulador aéreo con dos brazos antropomórficos diseñado para aplicarse en tareas de inspección y mantenimiento en entornos industriales de difícil acceso para operarios humanos. El robot consiste en una plataforma aérea multirrotor equipada con dos brazos antropomórficos ultraligeros, así como el sistema de control integrado de la plataforma y los brazos. Una de las principales características del manipulador es la flexibilidad mecánica proporcionada en todas las articulaciones, lo que aumenta la seguridad en las interacciones físicas con el entorno y la protección del propio robot. Para ello se ha introducido un compacto y simple mecanismo de transmisión por muelle entre el eje del servo y el enlace de salida. La estructura en aluminio de los brazos ha sido cuidadosamente diseñada de forma que los actuadores estén aislados frente a cargas radiales y axiales que los puedan dañar. El manipulador desarrollado ha sido validado a través de experimentos en base fija y en pruebas de vuelo en exteriores.

This paper presents techniques for GPS based autonomous navigation of heavy vehicles at high speed. The control system has two main functions: vehicle position estimation and generation of the steering commands for the vehicle to follow a given path autonomously. Position estimation is based on fusion of measurements from a carrier-phase differential GPS system and odometric sensors using fuzzy logic. A Takagi-Sugeno fuzzy controller is used for steering commands generation, to cope with different road geometry and vehicle velocity. The presented system has been implemented in a 13 tons truck, and fully tested in very demanding conditions, i.e. high velocity and large curvature variations in paved and unpaved roads.

This paper addresses the problem of collision avoidance in scenarios with multiple aerial vehicles and proposes a method based on a Legendre pseudospectral collocation in order to compute the solution trajectories and guarantee that the safety distance between them is always maintained. The method uses a rolling horizon policy in which trajectories are planned up to a given time horizon, thus considering a much smaller problem space. Then, the system is applied iteratively. Studies have been performed to set the values of the look-ahead time and the number of collocations points. The computational load and scalability of the method are also studied in randomly generated scenarios to test its application in real time. Experiments have been also carried out in the multivehicle aerial testbed of the Center for Advanced Aerospace Technologies (Seville, Spain)

This paper presents the design and experimental validation of a Fault Detection, Identification and Recovery (FDIR) system intended for multi-UAV applications. The system exploits the information provided by internal position, attitude and visual sensors onboard the UAVs of the fleet for detecting faults in the measurements of the position and attitude sensors of any of the member vehicles. Considering the observations provided by two or more UAVs in a cooperative way, it is possible to identify the source of the fault, but also implement a Cooperative Virtual Sensor (CVS) which provides a redundant position and velocity estimation of the faulty UAV that can be used for replacing its internal sensor. The vision-based FDIR system has been validated experimentally with quadrotors in an indoor testbed. In particular, fault detection and identification has been evaluated injecting a fault pattern offline on the position measurements, while the CVS has been applied in real time for the recovery phase.