Dissertations / Theses on the topic 'Robotic applications'

Thesis (Doctor of Philosophy)--Case Western Reserve University, 2010
Department of EECS - Computer Engineering Title from PDF (viewed on 2010-05-25) Includes abstract Includes bibliographical references and appendices Available online via the OhioLINK ETD Center

In this thesis a multilayer actuator structure is developed for the I-SWARM project. In order to build an energy efficient and low voltage actuator system for the 3x3x3 mm3 robot, the resonance drive mode in combination with a ten layer multilayer structure build upon the most suitable substrate material was used. Two different sizes of the locomotion module were built. The first one is five times larger than the small version. It has five active layers and is simpler to work with and to test. The small module has three 2x0.4 mm can-tilevers on a 3x3mm body and ten active layers.

The multilayer process involve easily defined flexible printed circuit (FPC) board as substrate material, spin coating piezoelectric poly(vinylidenefluoride- tetrafluoroethylene) P(VDF-TrFE) as active stack material, and evaporated aluminum electrodes on each active polymer layer. By using different shadow masks for each electrode layer, special inter con-tact areas can be contacted from above after the polymer has been removed by an Inductively Coupled Plasma (ICP) etch. The contours of the locomo-tion module was etched in a Reactive Ion Etch (RIE) equipment. Both the cupper layer in the FPC and the electrode layers of the active stack, were used as etch mask.

The deflections of the cantilevers were measured at low voltages to ensure as realistic drive voltage as possible for the I-SWARM robot. The large struc-ture showed a 250 µm deflection at 4 V and 170 Hz resulting in a Q-value of 19. The deflection of the small structure was 8 µm at 3.3 V and 5000 Hz and the measured Q-value was 31.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1990.
Title as it appears in the M.I.T. Graduate List, June, 1990: Machine vision for simulated spacecraft operations.
Includes bibliographical references (leaf 70).
by Ender St. John-Olcayto.
M.S.

Mestrado em Engenharia Electrónica e Telecomunicações
A procura e reconhecimento de padrões foi sempre um desafio para a mente humana e sem dúvida a sua maior capacidade. Esta tese encontra-se inserida no domínio do RoboCup e apresenta uma solução em tempo-real para a detecção de objectos através do processamento de imagem. Ao longo do trabalho, desenvolvemos vários algoritmos para análise de imagem com vista a encontrar objectos através da sua cor e das suas propriedades morfológicas. Nos algoritmos baseados na procura por cor, foram usados métodos de segmentação de cor e procura radial na imagem, permitindo bom desempenho em tempo-real. A pesquisa por propriedades morfológicas baseia-se em algoritmos de detecção de contornos em conjunto com a transformada circular de Hough. Ambos algoritmos, procura por cor e por características morfológicas, provaram a sua fiabilidade, sendo capazes de boas taxas de detecção em condições de tempo-real. Para além do anteriormente referido, foi desenvolvida uma biblioteca para manipular imagens e assegurar uma abstracção sobre os possíveis modos da imagem e uma ferramenta para ajudar na calibração da visão perspectiva. ABSTRACT: The search and recognition of patterns has always been a challenge for the human mind, and without any doubts its biggest capacity. This thesis is inserted in the RoboCup domain and presents a real-time solution to object detection using image analysis. In this work, we developed several image analysis algorithms to find objects based in their color and morphological properties. The color based search algorithms use color segmentation methods along with radial image scanning, allowing real-time performances. The morphological analysis is based in edge detection algorithms and the circular Hough transform. Both algorithms, search for color and morphological properties, proved their reliability, being capable of good detection ratios in real-time situations. Moreover, this thesis presents several tools, namely, an image library created to better manipulate the images and assure abstraction over the possible image modes acquired by digital cameras, and a tool to help in the perspective vision calibration.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 204-208).
This work investigates, defines, and expands on the use of robotic arms in digital fabrication, design, and art through methods including 3D printing, milling, sculpting, functionally graded fabrication, construction-scale additive manufacturing, jammable granular system design, light painting, and volumetric sensing. While most current applications of robotics in manufacturing rely on repetitive automation and assembly tasks, the flexibility, dexterity, and precision of industrial robotic arms provide for design opportunities of multi-functionary roles. Through exploration and demonstration, a multipurpose fabrication platform was developed using a KUKA KR5 sixx R850 robotic arm. The platform is capable of conventional manufacturing techniques spanning the three traditional fabrication categories: additive, subtractive, and formative. Case studies and digital design fabrication protocols were developed as part of the robotic platform to demonstrate these three types of fabrication including 3D printing, multi-axis milling, and clay sculpting, respectively. Compound processes, such as combining 3D printing and milling, were developed that offer product-, and process-based improvements over standalone techniques. The benefits and drawbacks of a multi-fabrication platform are discussed, including cost, physical footprint, resolution, and flexibility. In addition to replicating conventional manufacturing techniques with a single robotic platform, several novel applications were developed which take advantage of the flexibility of an arm system. First, functionally graded 3D printing was explored using concrete through which density gradients were shown to achieve higher structural efficiency. A novel construction-scale additive manufacturing process capable of 3D printing building components was developed. Secondly, direct recycling 3D printing was developed where waste thermoplastic products are transformed into feedstock and printed into new components within a single operation. Work conducted on jammed granular structures, where external pressure controls system stiffness and strength, resulted in several new formative fabrication possibilities. Combined with robotics, waste-free digital casting using jammable materials was enabled along with a variety of design projects including the design of robotic arms themselves. Finally, the use of robotic arms for fabrication of material and environmental properties without mechanical force transfer was explored. Coined immaterial fabrication,t his fabrication category captures methods that do not fall within the definitions of additive, subtractive, or formative processes. Work produced in this area includes a volumetric sensing technique and robotic light paintings that reveal thermal, electromagnetic, and optical fields.
by Steven J. Keating.
S.M.

Aquesta tesi s'emmarca dins del projecte CICYT TAP 1999-0443-C05-01. L'objectiu d'aquest projecte és el disseny, implementació i avaluació de robots mòbils, amb un sistema de control distribuït, sistemes de sensorització i xarxa de comunicacions per realitzar tasques de vigilància. Els robots han de poder-se moure per un entorn reconeixent la posició i orientació dels diferents objectes que l'envolten. Aquesta informació ha de permetre al robot localitzar-se dins de l'entorn on es troba per poder-se moure evitant els possibles obstacles i dur a terme la tasca encomanada. El robot ha de generar un mapa dinàmic de l'entorn que serà utilitzat per localitzar la seva posició. L'objectiu principal d'aquest projecte és aconseguir que un robot explori i construeixi un mapa de l'entorn sense la necessitat de modificar el propi entorn.
Aquesta tesi està enfocada en l'estudi de la geometria dels sistemes de visió estereoscòpics formats per dues càmeres amb l'objectiu d'obtenir informació geomètrica 3D de l'entorn d'un vehicle. Aquest objectiu tracta de l'estudi del modelatge i la calibració de càmeres i en la comprensió de la geometria epipolar. Aquesta geometria està continguda en el que s'anomena emph{matriu fonamental}. Cal realitzar un estudi del càlcul de la matriu fonamental d'un sistema estereoscòpic amb la finalitat de reduir el problema de la correspondència entre dos plans imatge. Un altre objectiu és estudiar els mètodes d'estimació del moviment basats en la geometria epipolar diferencial per tal de percebre el moviment del robot i obtenir-ne la posició. Els estudis de la geometria que envolta els sistemes de visió estereoscòpics ens permeten presentar un sistema de visió per computador muntat en un robot mòbil que navega en un entorn desconegut. El sistema fa que el robot sigui capaç de generar un mapa dinàmic de l'entorn a mesura que es desplaça i determinar quin ha estat el moviment del robot per tal de emph{localitzar-se} dins del mapa.
La tesi presenta un estudi comparatiu dels mètodes de calibració de càmeres més utilitzats en les últimes dècades. Aquestes tècniques cobreixen un gran ventall dels mètodes de calibració clàssics. Aquest mètodes permeten estimar els paràmetres de la càmera a partir d'un conjunt de punts 3D i de les seves corresponents projeccions 2D en una imatge. Per tant, aquest estudi descriu un total de cinc tècniques de calibració diferents que inclouen la calibració implicita respecte l'explicita i calibració lineal respecte no lineal. Cal remarcar que s'ha fet un gran esforç en utilitzar la mateixa nomenclatura i s'ha estandaritzat la notació en totes les tècniques presentades. Aquesta és una de les dificultats principals a l'hora de poder comparar les tècniques de calibració ja què cada autor defineix diferents sistemes de coordenades i diferents conjunts de paràmetres. El lector és introduït a la calibració de càmeres amb la tècnica lineal i implícita proposada per Hall i amb la tècnica lineal i explicita proposada per Faugeras-Toscani. A continuació es passa a descriure el mètode a de Faugeras incloent el modelatge de la distorsió de les lents de forma radial. Seguidament es descriu el conegut mètode proposat per Tsai, i finalment es realitza una descripció detallada del mètode de calibració proposat per Weng. Tots els mètodes són comparats tant des del punt de vista de model de càmera utilitzat com de la precisió de la calibració. S'han implementat tots aquests mètodes i s'ha analitzat la precisió presentant resultats obtinguts tant utilitzant dades sintètiques com càmeres reals.
Calibrant cada una de les càmeres del sistema estereoscòpic es poden establir un conjunt de restriccions geomètri ques entre les dues imatges. Aquestes relacions són el que s'anomena geometria epipolar i estan contingudes en la matriu fonamental. Coneixent la geometria epipolar es pot: simplificar el problema de la correspondència reduint l'espai de cerca a llarg d'una línia epipolar; estimar el moviment d'una càmera quan aquesta està muntada sobre un robot mòbil per realitzar tasques de seguiment o de navegació; reconstruir una escena per aplicacions d'inspecció, propotipatge o generació de motlles. La matriu fonamental s'estima a partir d'un conjunt de punts en una imatges i les seves correspondències en una segona imatge. La tesi presenta un estat de l'art de les tècniques d'estimació de la matriu fonamental. Comença pels mètode lineals com el dels set punts o el mètode dels vuit punts, passa pels mètodes iteratius com el mètode basat en el gradient o el CFNS, fins arribar las mètodes robustos com el M-Estimators, el LMedS o el RANSAC. En aquest treball es descriuen fins a 15 mètodes amb 19 implementacions diferents. Aquestes tècniques són comparades tant des del punt de vista algorísmic com des del punt de vista de la precisió que obtenen. Es presenten el resultats obtinguts tant amb imatges reals com amb imatges sintètiques amb diferents nivells de soroll i amb diferent quantitat de falses correspondències.
Tradicionalment, l'estimació del moviment d'una càmera està basada en l'aplicació de la geometria epipolar entre cada dues imatges consecutives. No obstant el cas tradicional de la geometria epipolar té algunes limitacions en el cas d'una càmera situada en un robot mòbil. Les diferencies entre dues imatges consecutives són molt petites cosa que provoca inexactituds en el càlcul de matriu fonamental. A més cal resoldre el problema de la correspondència, aquest procés és molt costós en quant a temps de computació i no és gaire efectiu per aplicacions de temps real. En aquestes circumstàncies les tècniques d'estimació del moviment d'una càmera solen basar-se en el flux òptic i en la geometria epipolar diferencial. En la tesi es realitza un recull de totes aquestes tècniques degudament classificades. Aquests mètodes són descrits unificant la notació emprada i es remarquen les
semblances i les diferencies entre el cas discret i el cas diferencial de la geometria epipolar. Per tal de poder aplicar aquests mètodes a l'estimació de moviment d'un robot mòbil, aquest mètodes generals que estimen el moviment d'una càmera amb sis graus de llibertat, han estat adaptats al cas d'un robot mòbil que es desplaça en una superfície plana. Es presenten els resultats obtinguts tant amb el mètodes generals de sis graus de llibertat com amb els adaptats a un robot mòbil utilitzant dades sintètiques i seqüències d'imatges reals.
Aquest tesi finalitza amb una proposta de sistema de localització i de construcció d'un mapa fent servir un sistema estereoscòpic situat en un robot mòbil. Diverses aplicacions de robòtica mòbil requereixen d'un sistema de localització amb l'objectiu de facilitar la navegació del vehicle i l'execució del les trajectòries planificades. La localització es sempre relativa al mapa de l'entorn on el robot s'està movent. La construcció de mapes en un entorn desconegut és una tasca important a realitzar per les futures generacions de robots mòbils. El sistema que es presenta realitza la localització i construeix el mapa de l'entorn de forma simultània. A la tesi es descriu el robot mòbil GRILL, que ha estat la plataforma de treball emprada per aquesta aplicació, amb el sistema de visió estereoscòpic que s'ha dissenyat i s'ha muntat en el robot. També es descriu tots el processos que intervenen en el sistema de localització i construcció del mapa. La implementació d'aquest processos ha estat possible gràcies als estudis realitzats i presentats prèviament (calibració de càmeres, estimació de la matriu fonamental, i estimació del moviment) sense els quals no s'hauria pogut plantejar aquest sistema. Finalment es presenten els mapes en diverses trajectòries realitzades pel robot GRILL en el laboratori.
Les principals contribucions d'aquest treball són:
·Un estat de l'art sobre mètodes de calibració de càmeres. El mètodes són comparats tan des del punt de vista del model de càmera utilitzat com de la precisió dels mètodes.
·Un estudi dels mètodes d'estimació de la matriu fonamental. Totes les tècniques estudiades són classificades i descrites des d'un punt de vista algorísmic.
·Un recull de les tècniques d'estimació del moviment d'una càmera centrat en el mètodes basat en la geometria epipolar diferencial. Aquestes tècniques han estat adaptades per tal d'estimar el moviment d'un robot mòbil.
·Una aplicació de robòtica mòbil per tal de construir un mapa dinàmic de l'entorn i localitzar-se per mitja d'un sistema estereoscòpic. L'aplicació presentada es descriu tant des del punt de vista del maquinari com del programari que s'ha dissenyat i implementat.
Human eyes have been widely studied by the scientific community so that its operation principle is widely known. Computer vision tries to copy the way human beings perceive visual information by means of using cameras acting as eyeballs and computers aspiring to process this information in an --intelligent way". The complex task of being conscious of reality is obviously divided into a set of simpler problems which covers from image acquisition to scene description. One of the main applications is robot perception in which a mobile robot is equipped with a computer vision system. Robots may be able to navigate around an unknown structured environment acquiring visual information of their surroundings with the aim of estimating the position and orientation of every obstacle. Moreover, the pose of the vehicle has to be estimated as accurate as possible. Hence, the motion of the vehicle might be also computed allowing the localization of the vehicle with respect to the 3D map.
This thesis is focused on the study of the geometry involved in stereo vision systems composed by two cameras with the aim of obtaining 3D geometric information of the vehicle surroundings. This objective deals to the study of camera modelling and calibration and the comprehension of the epipolar geometry. Then, the computation of the fundamental matrix of a stereoscopic system is surveyed with the aim of reducing the correspondence problem between both image planes. An accurate estimation of the fundamental matrix allows us not only to compute 3D information of the vehicle environments, but to validate it. Nevertheless, the traditional case of the epipolar geometry has some limitations in the common case of a single camera attached to a mobile robot. Disparities between two consecutive images are rather small at common image rates leading to numerical inaccuracies on the computation of the fundamental matrix. Then, another objective is the study of general vision-based egomotion estimation methods based on the differential epipolar constraint with the aim of perceiving the robot movement instead of its position.
The study of the geometry involved in stereo vision systems leads us to present a computer vision system mounted on a vehicle which navigates in an unknown environment. Two main tasks are faced: a) the localization of the vehicle; and b) the building of an absolute 3D map.
El sistema de visión humano ha sido ampliamente estudiado por la comunidad científica de forma que su principio de funcionamiento es profundamente conocido. La Visión por Computador trata de copiar la forma que nosotros los humanos percibimos la información visual por medio del uso de cámaras actuando como ojos y un ordenador aspirando a procesar toda la información de "forma inteligente". La compleja tarea de ser consciente de la realidad es obviamente dividida en un conjunto de problemas mucho más simples, los cuales abarcan des de la adquisición de la imagen a la descripción de la escena. Una de las numerosas aplicaciones es la percepción por parte de un robot, donde un robot móvil es equipado con un sistema informático de visión por computador. Estos robots deben ser capaces de navegar a lo largo de un entorno estructurado desconocido mediante la adquisición de información visual de su alrededor, con el objetivo de estimar la posición y orientación de todos los obstáculos. Además, la posición del vehículo debe ser estimada de la forma más precisa posible. De esta forma, el movimiento del vehículo puede ser también calculado lo que permite la localización del vehículo con respeto al mapa 3D.
Esta tesis profundiza en el estudio de la geometría existente en los sistemas de visión estéreo compuestos por dos cámaras con la intención de obtener información geométrica 3D del entorno del vehículo. Este objetivo lleva consigo la necesidad inicial de realizar un estudio de modelado de la cámara y calibración, y la compensación de la geometría epipolar. A continuación, el cálculo de la matriz fundamental de un sistema esteresocópico es analizado para reducir el problema de la correspondencia entre ambos planos de la imagen. Una estimación precisa de la matriz fundamental nos permite no solamente obtener la información 3D del entorno, sino también validar la misma. No obstante, la geometría epipolar tradicional sufre algunas limitaciones en el caso de una cámara montada en un robot móvil. La disparidad entre dos imágenes consecutivas es realmente mínima trabajando a velocidad estándar lo que conlleva a errores numéricos en el cálculo de la matriz fundamental. Por esta razón, otro objetivo es el estudio de los métodos de estimación del movimiento basados en la geometría epipolar diferencial con el objetivo de pervivir el movimiento del robot y su posición.
El estudio de la geometría inmersa en los sistemas de visión estéreo nos lleva a presentar un sistema de visión por computador montado en un vehículo capaz de navegar en un entorno desconocido. Dos tareas básicas son consideradas: a) la localización del vehículo; y b) la construcción de un mapa 3D absoluto.

Mestrado em Engenharia Eletrónica e Telecomunicações
The computer vision assumes an important relevance in the development of robotic applications. In several applications, robots need to use vision to detect objects, a challenging and sometimes difficult task. This thesis is focused on the study and development of algorithms to be used in detection and identification of objects on digital images to be applied on robots that will be used in practice cases. Three problems are addressed: Detection and identification of decorative stones for textile industry; Detection of the ball in robotic soccer; Detection of objects in a service robot, that operates in a domestic environment. In each case, different methods are studied and applied, such as, Template Matching, Hough transform and visual descriptors (like SIFT and SURF). It was chosen the OpenCv library in order to use the data structures to image manipulation, as well as other structures for all information generated by the developed vision systems. Whenever possible, it was used the implementation of the described methods and have been developed new approaches, both in terms of pre-processing algorithms and in terms of modification of the source code in some used functions. Regarding the pre-processing algorithms, were used the Canny edge detector, contours detection, extraction of color information, among others. For the three problems, there are presented and discussed experimental results in order to evaluate the best method to apply in each case. The best method for each application is already integrated or in the process of integration in the described robots.
A visão por computador assume uma importante relevância no desenvolvimento de aplicações robóticas, na medida em que há robôs que precisam de usar a visão para detetar objetos, uma tarefa desafiadora e por vezes difícil. Esta dissertação foca-se no estudo e desenvolvimento de algoritmos para a deteção e identificação de objetos em imagem digital para aplicar em robôs que serão usados em casos práticos. São abordados três problemas: Deteção e identificação de pedras decorativas para a indústria têxtil; Deteção da bola em futebol robótico; Deteção de objetos num robô de serviço, que opera em ambiente doméstico. Para cada caso, diferentes métodos são estudados e aplicados, tais como, Template Matching, transformada de Hough e descritores visuais (como SIFT e SURF). Optou-se pela biblioteca OpenCv com vista a utilizar as suas estruturas de dados para manipulação de imagem, bem como as demais estruturas para toda a informação gerada pelos sistemas de visão desenvolvidos. Sempre que possivel utilizaram-se as implementações dos métodos descritos tendo sido desenvolvidas novas abordagens, quer em termos de algoritmos de preprocessamento quer em termos de alteração do código fonte das funções utilizadas. Como algoritmos de pre-processamento foram utilizados o detetor de arestas Canny, deteção de contornos, extração de informação de cor, entre outros. Para os três problemas, são apresentados e discutidos resultados experimentais, de forma a avaliar o melhor método a aplicar em cada caso. O melhor método em cada aplicação encontra-se já integrado ou em fase de integração dos robôs descritos.

Tactile sensing is essential to the way humans and robots physically interact with the world. Humans use a complex system of nerves to touch and feel their surroundings. Researchers are attempting to achieve the same for robotics by studying and improving tactile sensing systems. Notably, tactile sensing for anthropomorphic robotics is of ongoing interest in homecare/ assistant, culinary, agriculture, and medical areas. The contribution to the development of tactile feedback systems can aid in advancing these robotic systems, allowing for more complex tasks to be performed. However, many of the available tactile sensing systems are currently very expensive and not practical for use in education and areas without access to expensive equipment. The price and complexity of available tactile sensing systems are currently challenging problems. This project investigates and explores sensing mechanisms, sensor structures, and fabrication methods to develop and evaluate several low-cost and accessible tactile sensors. Introducing novel designs and implementations of low-cost tactile sensors may bring more robotic systems closer to public application, positively affecting the funding and development of robotic advancements. Sequential implementation of varying materials and methods is performed to balance cost and complexity. This thesis reports: (1) An ultra low-cost graphite on paper sensor boasting an incredibly high response rate; (2) A low-cost and soft ionic liquid filled sensor capable of discerning directional pressures and temperature; (3) An easily integrable and flexible multifunctional pressure and temperature sensor; (4) A low-cost and easily integrable, flexible, and highly versatile multimodal pressure sensor. Evaluation of tactile sensing performance in robotic grippers successfully demonstrates the potential for these sensors in practical applications. This work has introduced and diversified existing technologies in robotic tactile sensing, improving the overall understanding and effectiveness of the materials and processes reported within. Presented works show the feasibility for ongoing research in areas that have been tapped into, furthering the development of smart and highly reliable low-cost tactile sensors.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Eng & Built Env
Science, Environment, Engineering and Technology
Full Text

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018.
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 143-151).
Fish present a natural source of inspiration for the design of high-performance under-water robots. Conventionally, fish-like robotic systems consist of a chain of rigid links connected by a series of rigid actuators. Devices of this nature have demonstrated impressive speeds and maneuverability, but from a practical perspective, their mechanical complexity can make them expensive to build and prone to failure. One possible solution is to replace the mechanical body of the swimmer with a passive elastic element. In this scenario, the robot uses a single actuator, housed within a rigid forebody, to generate a fish-like propagating wave along a flexible trailing tail. A number of groups have explored this approach, but so far, these devices have demonstrated relatively limited performance. Here, we study the kinematics and dynamics of elastic swimmers and apply the results of this process to guide the design and testing of a high-performance passive robotic swimmer. We begin the investigation with a first-principles approach. We use analytical models of fish hydrodynamics to characterize the kinematics of efficient propulsion in swimming animals. Armed with the insight developed through this process, we construct a numerical model of a passive elastic swimming sheet. Through the application of optimization methods, we demonstrate that the sheet can achieve 70-80% of the efficiency of an equivalent swimmer with actuators along its entire body. Based on this, we design, build, and test a passive elastic swimming robot which uses a novel inertia-based actuation system. Experiments with the robot show that it can achieve a top speed of 1m/s (3.17 body lengths/s) and a peak turning rate of 515 deg/s, among the highest reported to date, while swimming at efficiencies comparable to those of fully actuated systems.
by Alexander Joshua Wiens.
Ph. D.

Vast improvements in robotics and wireless communication have made teleoperated robots significantly more prevalent in industry, defense, and research. To help bridge the gap for these robots in the workplace, there has been a tremendous increase in research toward the development of biomimetic robotic hands that can simulate human operators. However, current methods of control are limited in scope and do not adequately represent human muscle memory and skills. The vision of this thesis is to provide a pathway for overcoming these limitations and open an opportunity for development and implementation of a cost effective methodology towards controlling a robotic hand. The first chapter describes the experiments conducted using Flexpoint bend sensors in conjunction with a simple voltage divider to generate a cost-effective data glove that is significantly less expensive than the commercially available alternatives. The data glove was able to provide sensitivity of less than 0.1 degrees. The second chapter describes the molding process for embedding pressure sensors in silicone skin and data acquisition from them to control the robotic hand. The third chapter describes a method for parsing and observing the information from the data glove and translating the relevant control variables to the robotic hand. The fourth chapter focuses on the feasibility of the brain computer interfaces (BCI) and successfully demonstrates the implementation of a simple brain computer interface in controlling a robotic hand.
Master of Science

Employees today have necessary daily tasks that do not require human handling. The objective is to investigate two processes if they are suitable for automation as well as to create and evaluate a prototype. The goals are to analyze the process, examine appropriate tools for automation, compare the tools, create and evaluate prototype, and perform an acceptance test. Robotic Process Automation is about automating tasks that humans have to do. Good candidates for automation are time-consuming, repetitive, rule-based tasks, prone to human er- rors with clear goals and expectations. The preliminary study was conducted in the form of a literature study of web-based sources, and the analysis was done by breaking down the process in different parts. The comparison was carried out by investigating the features of these tools. The prototype was created on Windows in UiPath tools and the robot will work on Internet Explorer and Excel, which will have a macro written in Visual Basic for Applications. The client will look at the criteria given and also on the prototype output and provide a subjective response. UiPath, Workfusion, and Selenium test programs were created. The prototype automatically logs on to Visma PX by entering username and password. Then it navigates, searches for an assignment and retrieves the data available. Indata is filtered and typed into Excel for each activity and employee. Finally, a macro creates graphs. Time tests show that UiPath is significantly more optimized and faster at completing the test programs. UiPath has strong benefits with its tools.
Anställda idag har nödvändiga vardagsuppgifter som inte kräver mänsklig inverkan och tanken är att frigöra dessa uppgifter. Projektets övergripande syfte är att undersöka två processer om de är lämpliga för automation samt att skapa och utvärdera en prototyp. Målen är att analysera processen, undersöka lämpliga verktyg för automatisering, jämföra verktygen, skapa en prototyp, utvärdera prototypen och utföra ett acceptanstest. Robotic Process Automation handlar om att automatisera uppgifter som människor gör. Bra kandidater för automatisering är tidskrävande, repetitiva, regelbaserade uppgifter, benägna till mänskliga fel med klara mål och förväntningar. Förstudien genomfördes i form av en litteraturstudie av webbaserade källor och analysen gjordes genom att bryta ner processen i olika delar. Jämförelsen genomfördes genom att undersöka de funk- tioner som verktygen har. Prototypen skapas på Windows i verktygen UiPath och roboten kommer att arbeta på Internet Explorer och mot Excel som kommer ha ett makro skrivet i Visual Basic for Applications. Beställaren kommer att titta på de kriterier som gavs och även på prototypens utdata och ge en subjektiv respons. Testprogrammen i UiPath, Workfusion och Selenium skapades med sina respektive funktioner. Prototypen loggar automatiskt in på Visma PX genom att skriva in användarnamn och lösenord. Sedan navigerar den i verktyget, söker på ett uppdrag och hämtar den data som finns. Indata filtreras och skrivs in i Excel för varje aktivitet och anställd. Slutligen körs ett makro som skapar grafer. Tidstesterna visar att UiPath är betydligt mer optimerad och snabbare på att slutföra testprogrammen. Jämförelserna visar att UiPath har starka fördelar med sitt verktyg.

Friction Stir Welding (FSW) is a new technology which joins materials by using frictional heat. Inthe first part of this thesis, a profound literature study is performed. The basic principles, therobotic implementation and possibilities to use FSW for high strength titanium alloys areexamined. In the next phase, a FSW-tool is modelled and implemented on an industrial robot in arobot simulation program. Reachability tests are carried out on car body parts and jet engineparts. By using a simulation program with embedded collision detection, all possible weldinglocations are determined on the provided parts. Adaptations like a longer FSW-tool and amodified design are suggested in order to get a better reachability. In different case studies, thenumber of required robots and the reduction of weight and time are investigated and comparedto the current spot welding process.

Thesis: S.M., Massachusetts Institute of Technology, School of Architecture and Planning, Program in Media Arts and Sciences, 2015.
"June 2015." Cataloged from PDF version of thesis.
Includes bibliographical references (pages 135-136).
The work of this thesis aims to enable the fast prototyping of multi-axis wearable robotic systems by developing a new modular electronics system. The flexible, scalable electronics architecture (FlexSEA) developed for this thesis fills the void between embedded systems used in commercial devices and in research prototypes. This system provides the required hardware and software for precise motion control, data acquisition, and networking. Scalability is obtained through the use of fast industrial communication protocols between the modules, and the standardization of the peripheral interfaces. Hardware and software encapsulation is used to provide high-performance, real-time control of the actuators while keeping the high-level control development fast, safe and simple. The FlexSEA kits are composed of two custom circuit boards (advanced brushless motor driver and microcontroller board), one commercial embedded computer, a complete software stack and documentation. During its development it has been integrated into a powered prosthetic knee as well as an autonomous ankle exoskeleton. To assess the usability of the FlexSEA kit, a new user successfully used a kit to read sensors and control an output device in less than three hours. FlexSEA simplifies and accelerates wearable robotics prototyping.
by Jean-François Duval.
S.M.

Current adhesives and gripping mechanisms used in many robotics applications function on very specific surface types or at defined attachment locations. A controllable, i.e. ON-OFF, adhesive mechanism that can operate on a wide range of surfaces would be very advantageous. Such a device would have applications ranging from robotic gripping and climbing to satellite docking and inspection/service missions. The main goal of the research presented here was to create such an attachment mechanism through the use of a new hybrid adhesive technology. The newly developed adhesive technology is a hybridization of electrostatic and micro-structured dry adhesion. The result provides enhanced robustness and utility, particularly on rough surfaces. There were challenges not only in the integration of these two adhesive elements but also with its application in a complete gripping mechanism.

Electrostatic and directional dry adhesives were both individually investigated. The electrode geometry for an electrostatic adhesive was optimized for maximum adhesion force using finite element analysis software. Optimization results were then verified through experimental testing. New manufacturing techniques were also developed for electrostatic adhesives that utilized a metalized mesh embedded in a silicone polymer and Kapton film based construction, greatly improving adhesion. The micro-structured dry adhesive used was provided by Dr. Aaron Parness, from the NASA Jet Propulsion Lab (JPL), and consists of an array of vertical stalks with an angled front face, referred to as micro-wedges. The hybrid electrostatic dry adhesive (EDA) was created by fabricating the electrostatic adhesive directly on top of a dry adhesive mold. This process created an array of dry adhesive micro-wedges directly on the surface of the electrostatic adhesive. In operation the electrostatic adhesive provides a normal force which serves to pull the dry adhesive into the surface substrate. With greater surface contact more of the dry adhesive is able to engage, bring the electrostatic adhesive even closer to the surface and increasing its effectiveness. Therefore, the combination of these two technologies creates a positive feedback cycle whose whole is often greater than the sum of its parts.

An interface mechanism is needed to transmit applied loads from a rigid structure to the flexible adhesive while still maintaining its conformability. This is especially important for strong adhesion on rough surfaces, such as tile and drywall. Different concepts such as a structured fibrillar hierarchy and a fluid-filled backing pouch have been explored. Additionally, finite element analysis was used to evaluate different fribrillar shapes and geometries for the structured hierarchy. The goal was to equalize the load distribution across the adhesive while still maintaining surface compliance. A gripper mechanism was also created which used a servo for actuation and three rigid tiles with a directional dry adhesive. It was tested on a perching Micro Air Vehicle (MAV) as well as in the RoboDome facility at NASA's Jet Propulsion lab to simulate a satellite docking/capture maneuver.

Background: With the pressing need for increased productivity and delivery of end products of uniform quality, industry is turning more and more to computer-based automation. At the present time, most of industrial automated manufacturing is carried out by specialpurpose machines, designed to perform specific functions in a manufacturing process. The inflexibility and generally high cost of these machines often referred to as hard automation systems, have led to a broad-based interest in the use of robots capable of performing a variety of manufacturing functions in a more flexible working environment and at lower production costs. A robot is a reprogrammable general-purpose manipulator with external sensors that can perform various assembly tasks. A robot may possess intelligence, which is normally due to computer algorithms associated with its controls and sensing systems. Industrial robots are general-purpose, computer-controlled manipulators consisting of several rigid links connected in series by revolute or prismatic joints. Most of today’s industrial robots, though controlled by mini and microcomputers are basically simple positional machines. They execute a given task by playing back a prerecorded or preprogrammed sequence of motion that has been previously guided or taught by the hand-held control teach box. Moreover, these robots are equipped with little or no external sensors for obtaining the information vital to its working environment. As a result robots are used mainly for relatively simple, repetitive tasks. More research effort has been directed in sensory feedback systems, which has resulted in improving the overall performance of the manipulator system. An example of a sensory feedback system would be: a vision Charge-Coupled Device (CCD) system. This can be utilized to manipulate the robot position dependant on the surrounding robot environment (various object profile sizes). This vision system can only be used within the robot movement envelope

This thesis documents the research that has led to the development, in prototype form, of a modular sensing system for use in a robot's work cell. The system implemented overcomes one of the major limitations of existing sensing systems, that is the difficulty of altering their sensing characteristics. The majority of sensing systems that are currently available are inflexible in that the addition of extra sensors requires, at the very least, substantial changes to both hardware and software. What these systems require is a facility through which users can easily, and readily, make changes to the configuration of the sensors they employ. In the modular system described, the sensors that provide the information about the robot's work cell are independent of the algorithms that make use of the information. That is, the sensors do not need any knowledge as to when or how the information they provide will be used. Similarly, the algorithms that make use of the data do not need any knowledge as to the provider of the data. This separation of data provider from data user enables the software that controls the sensors (and even the sensors themselves) to be upgraded without corresponding changes to the data user software. Additional sensors can easily be added to the system while redundant sensors can simply be removed. The location of objects within the robot's workspace is achieved by building a model of the workspace using the information provided by a number of sensors. As a prerequisite to model construction three problems had to be addressed. Firstly, the information extracted from different sensors is generally at different resolutions. Secondly, the representation of 3-D space requires large amounts of computer memory. Thirdly, the production of the 3-D model, particularly when a large number of sensors are involved requires a substantial amount of processor time. The first two problems were addressed using a data structure that allowed compact data storage, while the final problem was reduced by identifying parallel aspects of the processing and implementing them on a network of transputers. After the objects within the robot's workspace have been located, the next stage is to identify them. The identification is achieved by calculating the degree of match between measurable characteristics of the object to be identified and the same measurable characteristics of known objects. The degree of match, which is similar but not identical to the correlation function, between the object to be identified and each known object is then used to determine, if possible, the required identity of the object. The work contained within the thesis not only demonstrates the feasibility and benefits of a modular sensing system, over traditional sensing system, but has brought to light some points that will need further thought before a fully functional system is produced. The last chapter contains, in addition to a full and detailed list of conclusions made during the research, a summary of some of these areas that still require further work.

To solve the problem of limited tactile sensing in humanoid robotics as well as provide for future planned mechanical prostheses, an innovative tactile sensor system was created and embedded into two realistic-looking artificial skin gloves. These artificial skin tactile sensors used small piezoelectric ceramic disks to measure applied force at multiple points on each glove. The gloves were created using silicone rubber to simulate both the texture and look of human skin, while maintaining both flexibility and durability. The sensor outputs were buffered by high-impedance voltage-following operational amplifiers, and then read sequentially using a multiplexing scheme by a microcontroller. Sensor data were sent via USB to a computer, where a graphical user display was created to show the tactile information in real time. These prototypes successfully demonstrated the viability of small piezoelectric elements embedded in silicone rubber for use in creating flexible and elastic tactile sensors.

Image analysis, and its application to sensory input (computer vision) is a fairly mature field, so it is surprising that its techniques are not extensively used in robotic applications. The reason for this is that, traditionally, robots have been used in controlled environments where sophisticated computer vision was not necessary, for example in car manufacturing. As the field of robotics has moved toward providing general purpose robots that must function in the real world, it has become necessary that the robots be provided with robust sensors capable of understanding the complex world around them. However, when researchers apply techniques previously studied in image analysis literature to the field of robotics, several difficult problems emerge. In this thesis we examine four reasons why it is difficult to apply work in image analysis directly to real-time, general purpose computer vision applications. These are: improvement in the computational complexity of image analysis algorithms, robustness to dynamic and unpredictable visual conditions, independence from domain specific knowledge in object recognition and the development of debugging facilities. This thesis examines each of these areas making several innovative contributions in each area. We argue that, although each area is distinct, improvement must be made in all four areas before vision will be utilised as the primary sensory input for mobile, autonomous robotic applications. In the first area, the computational complexity of image analysis algorithms, we note the dependence of a large number of high-level processing routines on a small number of low-level algorithms. Therefore, improvement to a small set of highly utilised algorithms will yield benefits in a large number of applications. In this thesis we examine the common tasks of image segmentation, edge and straight line detection and vectorisation. In the second area, robustness to dynamic and unpredictable conditions, we examine how vision systems can be made more tolerant to changes of illumination in the visual scene. We examine the classical image segmentation task and present a method for illumination independence that builds on our work from the first area. The third area is the reliance on domain-specific knowledge in object recognition. Many current systems depend on a large amount of hard-coded domainspecific knowledge to understand the world around them. This makes the system hard to modify, even for slight changes in the environment, and very difficult to apply in a different context entirely. We present an XML-based language, the XML Object Definition (XOD) language, as a solution to this problem. The language is largely descriptive instead of imperative so, instead of describing how to locate objects within each image, the developer simply describes the properties of the objects. The final area is the development of support tools. Vision system programming is extremely difficult because large amounts of data are handled at a very fast rate. If the system is running on an embedded device (such as a robot) then locating defects in the code is a time consuming and frustrating task. Many development-support applications are available for specific applications. We present a general purpose development-support tool for embedded, real-time vision systems. The primary case study for this research is that of Robotic soccer, in the international RoboCup Four-Legged league. We utilise all of the research of this thesis to provide the first illumination-independent object recognition system for RoboCup. Furthermore we illustrate the flexibility of our system by applying it to several other tasks and to marked changes in the visual environment for RoboCup itself.

Image analysis, and its application to sensory input (computer vision) is a fairly mature field, so it is surprising that its techniques are not extensively used in robotic applications. The reason for this is that, traditionally, robots have been used in controlled environments where sophisticated computer vision was not necessary, for example in car manufacturing. As the field of robotics has moved toward providing general purpose robots that must function in the real world, it has become necessary that the robots be provided with robust sensors capable of understanding the complex world around them. However, when researchers apply techniques previously studied in image analysis literature to the field of robotics, several difficult problems emerge. In this thesis we examine four reasons why it is difficult to apply work in image analysis directly to real-time, general purpose computer vision applications. These are: improvement in the computational complexity of image analysis algorithms, robustness to dynamic and unpredictable visual conditions, independence from domain specific knowledge in object recognition and the development of debugging facilities. This thesis examines each of these areas making several innovative contributions in each area. We argue that, although each area is distinct, improvement must be made in all four areas before vision will be utilised as the primary sensory input for mobile, autonomous robotic applications. In the first area, the computational complexity of image analysis algorithms, we note the dependence of a large number of high-level processing routines on a small number of low-level algorithms. Therefore, improvement to a small set of highly utilised algorithms will yield benefits in a large number of applications. In this thesis we examine the common tasks of image segmentation, edge and straight line detection and vectorisation. In the second area, robustness to dynamic and unpredictable conditions, we examine how vision systems can be made more tolerant to changes of illumination in the visual scene. We examine the classical image segmentation task and present a method for illumination independence that builds on our work from the first area. The third area is the reliance on domain-specific knowledge in object recognition. Many current systems depend on a large amount of hard-coded domainspecific knowledge to understand the world around them. This makes the system hard to modify, even for slight changes in the environment, and very difficult to apply in a different context entirely. We present an XML-based language, the XML Object Definition (XOD) language, as a solution to this problem. The language is largely descriptive instead of imperative so, instead of describing how to locate objects within each image, the developer simply describes the properties of the objects. The final area is the development of support tools. Vision system programming is extremely difficult because large amounts of data are handled at a very fast rate. If the system is running on an embedded device (such as a robot) then locating defects in the code is a time consuming and frustrating task. Many development-support applications are available for specific applications. We present a general purpose development-support tool for embedded, real-time vision systems. The primary case study for this research is that of Robotic soccer, in the international RoboCup Four-Legged league. We utilise all of the research of this thesis to provide the first illumination-independent object recognition system for RoboCup. Furthermore we illustrate the flexibility of our system by applying it to several other tasks and to marked changes in the visual environment for RoboCup itself.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Information and Communication Technology
Full Text

Mobile robotics has become a rapidly developing field of interdisciplinary research within robotics. This promising field has attracted the attention of academicy, industry, several government agencies. Currently from security to personal service mobile robots are being used in a variety of tasks. The use of such robots is expected to only increase in the near future. In this study, it is aimed to design and manufacture a versatile robot base. This base is aimed to be the main driving unit for various applications performed both indoors and outdoors ranging from personal service and assistance to military applications. The study does not attempt to individually address any specific application, indeed it is aimed to shape up a robotic module that can be used in a wide range of application on different terrain with proper modification. The robot base is specifically designed for mixed terrain applications, yet this study attempts to provide some guidelines to help robot designers. The manufactured robot base is tested with tracks, wheels, and with both tracks and wheels, results are provided as guidelines to robot designers. Last but no the least, this study aims to obtain the know-how of building functional and flexible robots in Turkey by facilitating local resources as much as possible.

This study is a step forward in developing two different methods; one recognises human actions in an unbiased environment, the other predicts the next human action. The proposed methods that are based on deep learning, convolutional neural networks and long-short term memories, work regardless of camera motion, viewpoint variation, and irrelevant background context. The key outcome of this research is to enable an assistive robot to help a human peer performing an assembly task, using the proposed algorithms.

The domain of our work is mobile robotic task execution under dialogue control. In this domain, we examine some of the main issues related to automatic object identification based on a natural language description. In particular, the descriptions are referring expressions, and the goal is to identify the referent through a combination of dialogue and autonomous sensing.
The first step in this process involves bridging the gap between the user and system representations of the world. A number of concepts are expressed in natural language by vague terms like "the big box" and "very close to the door". We use fuzzy logic to map these vague terms onto the quantitative sensory data used in the computer representation of the environment. The uncertainty resulting from this mapping is abated by allowing the robotic agent to query the user for more information or to use sensors to collect missing data. We examine the trade-off between querying the user, sensing, and uncertainty in the situations where identification is not immediately realizable. Furthermore, we have devised an efficient questioning strategy based on the use of entropy to select the questions having the greatest discriminatory power over referent candidates. We have developed a framework to deal with each of these issues and have implemented a working system to demonstrate our strategies.

The use of robotic vehicles to detect and remove unexploded ordnance (UXO) from battlefields and training ranges is currently being explored by the Naval Explosive Ordnance Technology Division, Indian Head, Maryland. In support of this effort, research was conducted on the characterization and use of small commercially available magnetic inductance sensors to detect a variety of common U.S. submunitions. Sensor test bed mounting on a small wheeled vehicle with a sweep device allowed for dynamic testing against submunitions under laboratory and field conditions.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 23-25).
An experimental study was performed with the aim of developing a technique for manufacturing composite parts for use in dynamic robotic applications in lieu of heavy and expensive metal parts used in conventional robotic systems. There is already a wide usage of sandwich board materials in load bearing applications, but these do not provide equal strength in all directions, particularly compressive strength. Additionally, these materials are only available in two-dimensional shapes. The process developed over the course of this project seeks to make a fully covered composite of any desired geometries. The specific robotics project addressed was the hyper dynamic quadruped robotic platform, which ultimately seeks to design and construct a robot capable of a high speed gallop. This thesis began exploring methods of fabricating parts for one of the legs of the platform, specifically a radius part. Manufactured components needed to be both light in weight to facilitate ease of movement for the robot and strong enough to withstand the forces from the shifting weight during running. Proposed design parameters called for a foam core with a hard plastic shell to meet these needs. This technique can lead to a cheaper manufacturing method with a potential impact on the future robotics industry. After an investigation into the properties of different liquid polyurethane foams and plastics, the manufacturing techniques explored began with machining molds for both the inner core and outer shell of composite parts into wax blocks. The project aims were to develop a prototyping process, but this can lead to mass-production. Two versions of a manufacturing process with these blocks were developed, one which uses an open mold and one which uses a closed mold. Either method is viable for fabrication, with a preference for the open mold in parts with simple geometry and small thickness, and for the closed mold in larger parts or ones with complicated or interrupted outer perimeters.
by Theresa Dixon.
S.B.

Mestrado em Engenharia Mecânica
A tecnologia háptica está cada vez mais presente na sociedade em áreas tão diversas como a saúde, a robótica e o lazer. Nesta dissertação desenvolveuse um dispositivo háptico com o objetivo de ser utilizado na teleoperação robótica e na reabilitação. O dispositivo desenvolvido permite ao utilizador, através de uma interface gráfica, controlar uma rede de mini-motores vibratórios individualmente ou utilizando estímulos previamente definidos. O utilizador consegue assim definir que motor, ou que conjunto de motores, pretende activar e a intensidade dos mesmos. No desenvolvimento do sistema utilizou-se a arquitetura ROS e a arquitetura Qt para a implementação do sistema de comunicação e da interface gráfica, e a plataforma Arduino para o comando dos motores vibratórios. Para a funcionalidade do dispositivo na área da reabilitação foi desenvolvida uma luva/manga composta por dezasseis motores vibratórios dispostos em pontos estratégicos tendo em conta os nervos do ser humano. Todo o sistema foi testado e avaliado por pessoal médico e por uma amostra de voluntários com vários backgrounds. Uma análise realizada aos dados recolhidos mostrou que o método proposto foi bem sucedido.
The haptic technology is increasingly present in society in diverse areas such as health, robotics and recreation. In the present work, a haptic device with the objective of being used in robotic teleoperation and rehabilitation was developed. The developed device allows the user, through the use of a graphical interface, to control a network of mini vibration motors, either individually or using predefined stimuli. Therefore, the user can define which motor, or which group of motors, to turn on and its intensity. In the development of the system the platform ROS and platform Qt were used in the implementation of the communication system and the graphical interface, and the platform Arduino was used to do the command of the vibration motors. To the functionality of the device, a glove/sleeve composed by sixteen vibration motors displayed in strategical points according the human being dermatomes was developed. The entire system was tested and evaluated by clinicians and by a sample of volunteers with different backgrounds. An analysis to the collected data showed that the proposed method was successful.

Using robots to perform the vision-based coverage tasks such as inspection, shape reconstruction and surveillance has attracted increasing attentions from both academia and industry in past few years. These tasks require the robot to carry vision sensors such as RGB camera, laser scanner, thermal camera to take surface measurement of the desired target objects, with a required surface coverage constraint. Due to the high demands and repetitive natures of these tasks, automatically generating efficient robotic motion plans could significantly reduce cost and improve productivity, which is highly desirable. Several planning approaches have been proposed for the vision-based coverage planning problems with robots in the past. However, these planning methods either only focused on coverage problems in 2D environment, or found less optimal results, or were specific to limited scenarios. In this thesis, we proposed the novel planning algorithms for vision-based coverage planning problems with industrial manipulators and Unmanned Aerial Vehicles (UAVs) in complex 3D environment. Different sampling and optimization methods have been used in the proposed planning algorithms to achieve better planning results. The very first and important step of these coverage planning tasks is to identify a suitable viewpoint set that satisfies the application requirements. This is considered as a view planning problem. The second step is to plan collision-free paths, as well as the visiting sequence of the viewpoints. This step can be formulated as a sequential path planning problem, or path planning problem for short. In this thesis, we developed view planning methods that generate candidate viewpoints using randomized sampling-based and Medial Object-based methods. The view planning methods led to better results with fewer required viewpoints and higher coverage ratios. Moreover, the proposed view planning methods were also applied to practical application in which the detailed 3D building model needs to be reconstructed when only 2D public map data is available. In addition to the proposed view planning algorithms, we also combined the view planning and path planning problems as a single coverage planning problem; and solved the combined problem in a single optimization process to achieve better results. The proposed planning method was applied to industrial shape inspection application with robotic manipulators. Additionally, we also extended the planning method to a industrial robotic inspection system with kinematic redundancy to enlarge the workspace and to reduce the required inspection time. Moreover, a learning-based robotic calibration method was also developed in order to accurately position vision sensors to desired viewpoints in these instances with industrial manipulators.

Autonomous navigation systems which operate in unknown or partially known environments strongly rely on sensor data to estimate the world state and do action planning. Considering such a scenario, the accuracy of the information supplied to the system is vital to make it able to operate correctly and safely. As a specific field of intelligent systems --in robotic applications-- it is desired to have semantic information of the environment for better recognition and navigation. For semantic perception, classification of objects into multiple classes is a fundamental requirement. On the other hand, it is an error-prone process as a consequence of variability of the scenes a robot may visit. However the lack of reliability of the classification step is often ignored or not explicitly measured. This problem is disregarded in existing literature. This master thesis work introduces a method to include uncertainty in the system. To this purpose, first the classifier and the specific dataset using an autonomous robot are created. Then the world scene is segmented and this information is projected to the 3D map. Later the uncertainty is estimated at different locations of the map. This map can then serve as input for variety of tasks such as exploration, active learning and human-robot interaction. Results on specific dataset generated using images from environment of TUM Theresienstrasse Campus show that uncertainty of classification is higher in misclassified areas.

Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2006.
Ashraf Saad, Committee Member ; Spyros Reveliotis, Committee Member ; Anthony Yezzi, Committee Member ; Erik Verriest, Committee Member ; Yorai Wardi, Committee Co-Chair ; Magnus Egerstedt, Committee Chair.

The implementation and running of physical security systems is costly and potentially hazardous for those employed to patrol areas of interest. From a technial perspective, the physical security problem can be seen as minimising the probability that intruders and other anomalous events will occur unobserved. A robotic solution is proposed using an artificial immune system, traditionally applied to software security, to identify threats and hazards: the dendritic cell algorithm. It is demonstrated that the migration from the software world to the hardware world is achievable for this algorithm and key properties of the resulting system are explored empirically and theoretically. It is found that the algorithm has a hitherto unknown frequency-dependent component, making it ideal for filtering out sensor noise. Weaknesses of the algorithm are also discovered, by mathematically phrasing the signal processing phase as a collection of linear classifiers. It is concluded that traditional machine learning approaches are likely to outperform the implemented system in its current form. However, it is also observed that the algorithm’s inherent filtering characteristics make modification, rather than rejection, the most beneficial course of action. Hybridising the dendritic cell algorithm with more traditional machine learning techniques, through the introduction of a training phase and using a non-linear classification phase is suggested as a possible future direction.

This thesis investigates the fusion of 3D visual information with 2D image cues to provide 3D semantic maps of large-scale environments in which a robot traverses for robotic applications. A major theme of this thesis was to exploit the availability of 3D information acquired from robot sensors to improve upon 2D object classification alone. The proposed methods have been evaluated on several indoor and outdoor datasets collected from mobile robotic platforms including a quadcopter and ground vehicle covering several kilometres of urban roads.