Ground truth determination for segmentation of tomographic volumes using interpolation

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

eVentos 2 - Autonomous sailboat control

Dissertação para obtenção do Grau de Mestre em Engenharia Electrotécnica e de Computadores

A volumetric hybrid representation for obstacle detection in all-terrain robots

This dissertation presents an approach aimed at three-dimensional perception’s obstacle detection on all-terrain robots. Given the huge amount of acquired information, the adversities such environments present to an autonomous system and the swiftness, thus required, from each of its navigation decisions, it becomes imperative that the 3-D perceptional system to be able to map obstacles and passageways in the most swift and detailed manner. In this document, a hybrid approach is presented bringing the best of several methods together, combining the lightness of lesser meticulous analyses with the detail brought by more thorough ones. Realizing the former, a terrain’s slope mapping system upon a low resolute volumetric representation of the surrounding occupancy. For the latter’s detailed evaluation, two novel metrics were conceived to discriminate the little depth discrepancies found in between range scanner’s beam distance measurements. The hybrid solution resulting from the conjunction of these two representations provides a reliable answer to traversability mapping and a robust discrimination of penetrable vegetation from that constituting real obstructions. Two distinct robotic platforms offered the possibility to test the hybrid approach on very different applications: a boat, under an European project, the ECHORD Riverwatch, and a terrestrial four-wheeled robot for a national project, the Introsys Robot.

Probabilistic constraint reasoning with Monte Carlo integration to robot localization

This work studies the combination of safe and probabilistic reasoning through the hybridization of Monte Carlo integration techniques with continuous constraint programming. In continuous constraint programming there are variables ranging over continuous domains (represented as intervals) together with constraints over them (relations between variables) and the goal is to find values for those variables that satisfy all the constraints (consistent scenarios). Constraint programming “branch-and-prune” algorithms produce safe enclosures of all consistent scenarios. Special proposed algorithms for probabilistic constraint reasoning compute the probability of sets of consistent scenarios which imply the calculation of an integral over these sets (quadrature). In this work we propose to extend the “branch-and-prune” algorithms with Monte Carlo integration techniques to compute such probabilities. This approach can be useful in robotics for localization problems. Traditional approaches are based on probabilistic techniques that search the most likely scenario, which may not satisfy the model constraints. We show how to apply our approach in order to cope with this problem and provide functionality in real time.

Teleoperation of a service robot using a mobile device

Teleoperation is a concept born with the rapid evolution of technology, with an intuitive meaning "operate at a distance." The first teleoperation system was created in the mid 1950s, which were handled chemicals. Remote controlled systems are present nowadays in various types of applications. This dissertation presents the development of a mobile application to perform the teleoperation of a mobile service robot. The application integrates a distributed surveillance (the result of a research project QREN) and led to the development of a communication interface between the robot (the result of another QREN project) and the vigilance system. It was necessary to specify a communication protocol between the two systems, which was implemented over a communication framework 0MQ (Zero Message Queue). For the testing, three prototype applications were developed before to perform the test on the robot.

AGNI: an API for the control of automomous service robots

With the continuum growth of Internet connected devices, the scalability of the protocols used for communication between them is facing a new set of challenges. In robotics these communications protocols are an essential element, and must be able to accomplish with the desired communication. In a context of a multi-­‐‑agent platform, the main types of Internet communication protocols used in robotics, mission planning and task allocation problems will be revised. It will be defined how to represent a message and how to cope with their transport between devices in a distributed environment, reviewing all the layers of the messaging process. A review of the ROS platform is also presented with the intent of integrating the already existing communication protocols with the ServRobot, a mobile autonomous robot, and the DVA, a distributed autonomous surveillance system. This is done with the objective of assigning missions to ServRobot in a security context.

Context classification for service robots

This dissertation presents a solution for environment sensing using sensor fusion techniques and a context/environment classification of the surroundings in a service robot, so it could change his behavior according to the different rea-soning outputs. As an example, if a robot knows he is outdoors, in a field environment, there can be a sandy ground, in which it should slow down. Contrariwise in indoor environments, that situation is statistically unlikely to happen (sandy ground). This simple assumption denotes the importance of context-aware in automated guided vehicles.

Non-cell autonomous neuroprotective effect of carbon monoxide : microglia-to-neuron communication

RESUMO: A isquémia cerebral é uma das doenças mais predominantes a nivel mundial, sendo uma das principais causas de mortalidade e invalidez. Parte da propagação de dano no cérebro é causado por inflamação descontrolada, causada principalmente por disfunção da microglia. Desta forma, existe a necessidade de tentar desenvolver estratégias para melhor compreender e modular as acções destas células. O monóxido de carbono (CO), é uma molécula endógena com provas dadas como anti-neuroinflamatório em vários modelos. Assim, o principal objectivo do trabalho foi o estudo do CO como um modulador da acção da microglia, com principal foco dado à comunicação entre estas células e neurónios, tentando entender se existe um efeito neuroprotector por inibição da inflamação. Um protocolo de meio condicionado foi estabelecido usando as linhas celulares BV2 e SH-SY5Y, de microglia e neurónio. A molécula CORM-A1, que liberta expontaniamente CO, foi usada como método de entrega da molécula às celulas. Demonstrámos que o pre-tratamento de células BV2 com CORM-A1 gera neuroprotecção já que reduz a morte celular de neurónios SH-SY5Y quando são incubados com meio condicionado de microglia activada em conjunto com o pró-oxidante t-BHP (tert-butil hidroperóxido). Assim, considerámos que o CO promove neuroprotecção ao inibir as acções inflamatórias da microglia. O papel anti-inflamatório da molécula CORM-A1 foi confirmado quando se verificou que pré-tratamento desta molécula em microglia BV2 limita a secreção de TNF-α mas estimula a secreção de IL-10. Por último, a CORM-A1 induziu a expressão do receptor da microglia CD200R1, molécula que participa na comunicação neurónio-microglia e fundamental para a modulação das acções inflamatórias destas últimas. Em suma, o nosso trabalho reforçou as propriedades anti-neuroinflamatórias do CO e uma capacidade de modular viabilidade neuronal através do seu efeito a nível de comunicação célula-célula. ---------------------------- ABSTRACT: Brain ischemia is a widespread disease worldwide, being one of the main causes of mortality and permanent disability. A portion of the damage that ensues following the ischemic event is caused by unrestrained inflammation, which is mainly orchestrated by exacerbated microglial activity. Hence, developing strategies for modulating microglial inflammation is a major concern nowadays. The endogenous molecule carbon monoxide (CO) has been shown to possess anti-neuroinflammatory properties using in vitro and in vivo approaches. Thus, our objective was to study CO as modulator of microglial activity, in particular in what concerns their communication with neurons, by promoting neuronal viability and limiting inflammatory output of activated microglia. A conditioned media strategy was established with BV2 microglia and SH-SY5Y neurons as cell models. CO-releasing molecule A1 (CORM-A1), a compound that releases CO spontaneously, was used as method of CO delivery to cells. We found that CORM-A1 pre-treatment in BV2 cells yields neuroprotective results, as it limits cell death when SH-SY5Y neurons are challenged with conditioned media from LPS-activated microglia and the pro-oxidant t-BHP (tert-butyl-hydroperoxide). Thus, we assumed carbon monoxide promotes neuroprotection via inhibition of microglial inflammation, displaying a non-cell autonomous role. CORM-A1 pre-treatment limited inflammation by inhibiting BV2 secretion of TNF-α and stimulating IL-10 production. These results reinforce that CO’s anti-inflammatory role confers neuroprotection, as the alterations in these cytokines occur concurrently with the increase in SH-SY5Y viability. Finally, we showed for the first time that carbon monoxide promotes the expression of CD200R1, a microglial receptor involved in neuron-glia communication and modulation of microglia inflammation. Further studies are necessary to clarify this role. Altogether, other than just highlighting CO as an anti-inflammatory and neuroprotective molecule, this work set the foundation for disclosing its involvement in cell-to-cell communication.

The potential field method and the nonlinear attractor dynamics approach: what are the differences?

One of the most popular approaches to path planning and control is the potential field method. This method is particularly attractive because it is suitable for on-line feedback control. In this approach the gradient of a potential field is used to generate the robot's trajectory. Thus, the path is generated by the transient solutions of a dynamical system. On the other hand, in the nonlinear attractor dynamic approach the path is generated by a sequence of attractor solutions. This way the transient solutions of the potential field method are replaced by a sequence of attractor solutions (i.e., asymptotically stable states) of a dynamical system. We discuss at a theoretical level some of the main differences of these two approaches.

RoboCup Soccer Leagues

RoboCup was created in 1996 by a group of Japanese, American, and European Artificial Intelligence and Robotics researchers with a formidable, visionary long-term challenge: “By 2050 a team of robot soccer players will beat the human World Cup champion team.” At that time, in the mid 90s, when there were very few effective mobile robots and the Honda P2 humanoid robot was presented to a stunning public for the first time also in 1996, the RoboCup challenge, set as an adversarial game between teams of autonomous robots, was fascinating and exciting. RoboCup enthusiastically and concretely introduced three robot soccer leagues, namely “Simulation,” “Small-Size,” and “Middle-Size,” as we explain below, and organized its first competitions at IJCAI’97 in Nagoya with a surprising number of 100 participants [RC97]. It was the beginning of what became a continously growing research community. RoboCup established itself as a structured organization (the RoboCup Federation www.RoboCup.org). RoboCup fosters annual competition events, where the scientific challenges faced by the researchers are addressed in a setting that is attractive also to the general public. and the RoboCup events are the ones most popular and attended in the research fields of AI and Robotics.RoboCup further includes a technical symposium with contributions relevant to the RoboCup competitions and beyond to the general AI and robotics.

[Preface to] RoboCup 2007: Robot Soccer World Cup XI

[Excerpt] The 11th RoboCup International Symposium was held during July 9–10, 2007 at the Fox Theatre in Atlanta, GA, immediately after the 2007 Soccer, Rescue and Junior Competitions. The RoboCup community has observed an increasing interest from other communities over the past few years, e.g., the robotics community.RoboCupisseenasasignificantapproachtotheevaluationofnewlydeveloped methods to many difficult problems in robotics. Atlanta was also the location of a RoboCup@Space demonstration, which reflected the role of AI and robotics in space exploration. Prior to the symposium, space agencies had expressed an interest in cooperating with RoboCup. A first step in this direction was a successful demonstration at RoboCup 2007, which was accompanied with aninvitedtalkgivenbyaleadingscientistfromtheJapanAerospaceExploration Agency JAXA. [...]

Development of autonomous maintenance in a furniture company

This article presents a work performed in the maintenance department of a furniture company in Portugal, in order to develop and implement autonomous maintenance. The main objective of the project was related to the objective to increase and make effective the autonomous maintenance tasks performed by production operators, and in this way avoiding unplanned downtime due to equipment failures. Although some autonomous maintenance tasks were already carried out within the company, a preliminary study revealed weaknesses in the application of this tool. In the initial phase of this pilot project, the main problems encountered at the level of autonomous maintenance were related to the lack of time to carry out these tasks, showing that the stipulated procedures were far from the real needs of the company. To solve these problems a pilot project was conducted, making several changes in the performance of autonomous maintenance tasks, making them standard and adapted to reality of each production line. There was a general improvement in the factory indicators, and essentially there was a behavioral change, since the operators felt that their opinions were taking into account and began to understand the importance of small tasks performed by them.

Structure and composition of the ground-herb community in a terra-firme Central Amazonian forest

The herb community of tropical forests is very little known, with few studies addressing its structure quantitatively. Even with this scarce body of information, it is clear that the ground herbs are a rich group, comprising 14 to 40% of the species found in total species counts in tropical forests. The present study had the objective of increasing the knowledge about the structure and composition of the ground-herb community and to compare the sites for which there are similar studies. The study was conducted in a tropical non-inundated and evergreen forest 90 km north of Manaus, AM. Ground herbs were surveyed in 22 transects of 40 m², distributed in five plots of 4 ha. The inventoried community was composed of 35 species, distributed in 24 genera and 18 families. Angiosperms were represented by 8 families and Pteridophytes by 10 families. Marantaceae (12 sp) and Cyperaceae (4 sp) were the richest families. Marantaceae and Poaceae were the families with greatest abundance and cover. Marantaceae, Poaceae, Heliconiaceae and Pteridophytes summed 96% of total herb cover, and therefore were responsible for almost all the cover of the community. The 10 most important species had 83.7% of the individuals. In general, the most abundant species were also the most frequent. Richness per transect varied from 7 to 19 species, and abundance varied from 30 to 114 individuals. The community structure was quite similar to 3 other sites in South America and one site in Asia.

Off-line simulation inspires insight: a neurodynamics approach to efficient robot task learning

There is currently an increasing demand for robots able to acquire the sequential organization of tasks from social learning interactions with ordinary people. Interactive learning-by-demonstration and communication is a promising research topic in current robotics research. However, the efficient acquisition of generalized task representations that allow the robot to adapt to different users and contexts is a major challenge. In this paper, we present a dynamic neural field (DNF) model that is inspired by the hypothesis that the nervous system uses the off-line re-activation of initial memory traces to incrementally incorporate new information into structured knowledge. To achieve this, the model combines fast activation-based learning to robustly represent sequential information from single task demonstrations with slower, weight-based learning during internal simulations to establish longer-term associations between neural populations representing individual subtasks. The efficiency of the learning process is tested in an assembly paradigm in which the humanoid robot ARoS learns to construct a toy vehicle from its parts. User demonstrations with different serial orders together with the correction of initial prediction errors allow the robot to acquire generalized task knowledge about possible serial orders and the longer term dependencies between subgoals in very few social learning interactions. This success is shown in a joint action scenario in which ARoS uses the newly acquired assembly plan to construct the toy together with a human partner.

Global vs. local nonlinear optimization techniques for human-like movement of an anthropomorphic robot

In this paper a comparison between using global and local optimization techniques for solving the problem of generating human-like arm and hand movements for an anthropomorphic dual arm robot is made. Although the objective function involved in each optimization problem is convex, there is no evidence that the admissible regions of these problems are convex sets. For the sequence of movements for which the numerical tests were done there were no significant differences between the optimal solutions obtained using the global and the local techniques. This suggests that the optimal solution obtained using the local solver is indeed a global solution.