Gradient-based reinforcement learning techniques for underwater robotics behavior learning

Darrerament, l'interès pel desenvolupament d'aplicacions amb robots submarins autònoms (AUV) ha crescut de forma considerable. Els AUVs són atractius gràcies al seu tamany i el fet que no necessiten un operador humà per pilotar-los. Tot i això, és impossible comparar, en termes d'eficiència i flexibilitat, l'habilitat d'un pilot humà amb les escasses capacitats operatives que ofereixen els AUVs actuals. L'utilització de AUVs per cobrir grans àrees implica resoldre problemes complexos, especialment si es desitja que el nostre robot reaccioni en temps real a canvis sobtats en les condicions de treball. Per aquestes raons, el desenvolupament de sistemes de control autònom amb l'objectiu de millorar aquestes capacitats ha esdevingut una prioritat. Aquesta tesi tracta sobre el problema de la presa de decisions utilizant AUVs. El treball presentat es centra en l'estudi, disseny i aplicació de comportaments per a AUVs utilitzant tècniques d'aprenentatge per reforç (RL). La contribució principal d'aquesta tesi consisteix en l'aplicació de diverses tècniques de RL per tal de millorar l'autonomia dels robots submarins, amb l'objectiu final de demostrar la viabilitat d'aquests algoritmes per aprendre tasques submarines autònomes en temps real. En RL, el robot intenta maximitzar un reforç escalar obtingut com a conseqüència de la seva interacció amb l'entorn. L'objectiu és trobar una política òptima que relaciona tots els estats possibles amb les accions a executar per a cada estat que maximitzen la suma de reforços totals. Així, aquesta tesi investiga principalment dues tipologies d'algoritmes basats en RL: mètodes basats en funcions de valor (VF) i mètodes basats en el gradient (PG). Els resultats experimentals finals mostren el robot submarí Ictineu en una tasca autònoma real de seguiment de cables submarins. Per portar-la a terme, s'ha dissenyat un algoritme anomenat mètode d'Actor i Crític (AC), fruit de la fusió de mètodes VF amb tècniques de PG.

A formalization for multi-agent decision support in cooperative environments. A framework for situated agents

La tesis propone un marco de trabajo para el soporte de la toma de decisiones adecuado para soportar la ejecución distribuida de acciones cooperativas en entornos multi-agente dinámicos y complejos. Soporte para la toma de decisiones es un proceso que intenta mejorar la ejecución de la toma de decisiones en escenarios cooperativos. Este proceso ocurre continuamente en la vida diaria. Los humanos, por ejemplo, deben tomar decisiones acerca de que ropa usar, que comida comer, etc. En este sentido, un agente es definido como cualquier cosa que está situada en un entorno y que actúa, basado en su observación, su interpretación y su conocimiento acerca de su situación en tal entorno para lograr una acción en particular.Por lo tanto, para tomar decisiones, los agentes deben considerar el conocimiento que les permita ser consientes en que acciones pueden o no ejecutar. Aquí, tal proceso toma en cuenta tres parámetros de información con la intención de personificar a un agente en un entorno típicamente físico. Así, el mencionado conjunto de información es conocido como ejes de decisión, los cuales deben ser tomados por los agentes para decidir si pueden ejecutar correctamente una tarea propuesta por otro agente o humano. Los agentes, por lo tanto, pueden hacer mejores decisiones considerando y representando apropiadamente tal información. Los ejes de decisión, principalmente basados en: las condiciones ambientales, el conocimiento físico y el valor de confianza del agente, provee a los sistemas multi-agente un confiable razonamiento para alcanzar un factible y exitoso rendimiento cooperativo.Actualmente, muchos investigadores tienden a generar nuevos avances en la tecnología agente para incrementar la inteligencia, autonomía, comunicación y auto-adaptación en escenarios agentes típicamente abierto y distribuidos. En este sentido, esta investigación intenta contribuir en el desarrollo de un nuevo método que impacte tanto en las decisiones individuales como colectivas de los sistemas multi-agente. Por lo tanto, el marco de trabajo propuesto ha sido utilizado para implementar las acciones concretas involucradas en el campo de pruebas del fútbol robótico. Este campo emula los juegos de fútbol real, donde los agentes deben coordinarse, interactuar y cooperar entre ellos para solucionar tareas complejas dentro de un escenario dinámicamente cambiante y competitivo, tanto para manejar el diseño de los requerimientos involucrados en las tareas como para demostrar su efectividad en trabajos colectivos. Es así que los resultados obtenidos tanto en el simulador como en el campo real de experimentación, muestran que el marco de trabajo para el soporte de decisiones propuesto para agentes situados es capaz de mejorar la interacción y la comunicación, reflejando en un adecuad y confiable trabajo en equipo dentro de entornos impredecibles, dinámicos y competitivos. Además, los experimentos y resultados también muestran que la información seleccionada para generar los ejes de decisión para situar a los agentes, es útil cuando tales agentes deben ejecutar una acción o hacer un compromiso en cada momento con la intención de cumplir exitosamente un objetivo colectivo. Finalmente, algunas conclusiones enfatizando las ventajas y utilidades del trabajo propuesto en la mejora del rendimiento colectivo de los sistemas multi-agente en situaciones tales como tareas coordinadas y asignación de tareas son presentadas.

Contribució al posicionament dinàmic de robots mòbils per mitjà d'un sistema làser

Aquesta tesi tracta el problema del posicionament de robots mòbils quan, en el decurs del moviment, es realitzen mesures angulars relatives al robot de l'orientació de la recta entre un dels seus punts i punts de l'entorn de posició coneguda. Es considera que les mesures angulars són fetes per un sensor làser giratori que detecta diferents reflectors catadiòptrics fixos. La contribució principal és el desenvolupament d'un algorisme dinàmic, basat en un filtre de Kalman estès (EKF), que estima a cada instant de temps l'estat format pels angles associats als reflectors. La simulació hodomètrica dels angles entre mesures directes del sensor làser garanteix l'ús consistent i continuat dels mètodes de triangulació per a determinar la posició i l'orientació del robot. Inclou simulacions informàtiques i experiments per a validar la precisió del mètode de posicionament proposat. En l'experimentació s'utilitza un robot mòbil omnidireccional amb tres rodes de lliscament direccional de corrons esfèrics.

On intentional and social agents with graded attitudes

La principal contribución de esta Tesis es la propuesta de un modelo de agente BDI graduado (g-BDI) que permita especificar una arquitetura de agente capaz de representar y razonar con actitudes mentales graduadas. Consideramos que una arquitectura BDI más exible permitirá desarrollar agentes que alcancen mejor performance en entornos inciertos y dinámicos, al servicio de otros agentes (humanos o no) que puedan tener un conjunto de motivaciones graduadas. En el modelo g-BDI, las actitudes graduadas del agente tienen una representación explícita y adecuada. Los grados en las creencias representan la medida en que el agente cree que una fórmula es verdadera, en los deseos positivos o negativos permiten al agente establecer respectivamente, diferentes niveles de preferencias o de rechazo. Las graduaciones en las intenciones también dan una medida de preferencia pero en este caso, modelan el costo/beneficio que le trae al agente alcanzar una meta. Luego, a partir de la representación e interacción de estas actitudes graduadas, pueden ser modelados agentes que muestren diferentes tipos de comportamiento. La formalización del modelo g-BDI está basada en los sistemas multi-contextos. Diferentes lógicas modales multivaluadas se han propuesto para representar y razonar sobre las creencias, deseos e intenciones, presentando en cada caso una axiomática completa y consistente. Para tratar con la semántica operacional del modelo de agente, primero se definió un calculus para la ejecución de sistemas multi-contextos, denominado Multi-context calculus. Luego, mediante este calculus se le ha dado al modelo g-BDI semántica computacional. Por otra parte, se ha presentado una metodología para la ingeniería de agentes g-BDI en un escenario multiagente. El objeto de esta propuesta es guiar el diseño de sistemas multiagentes, a partir de un problema del mundo real. Por medio del desarrollo de un sistema recomendador en turismo como caso de estudio, donde el agente recomendador tiene una arquitectura g-BDI, se ha mostrado que este modelo es valioso para diseñar e implementar agentes concretos. Finalmente, usando este caso de estudio se ha realizado una experimentación sobre la flexibilidad y performance del modelo de agente g-BDI, demostrando que es útil para desarrollar agentes que manifiesten conductas diversas. También se ha mostrado que los resultados obtenidos con estos agentes recomendadores modelizados con actitudes graduadas, son mejores que aquellos alcanzados por los agentes con actitudes no-graduadas.

Studies of dynamics of physical agent ecosystems

This thesis addresses the problem of learning in physical heterogeneous multi-agent systems (MAS) and the analysis of the benefits of using heterogeneous MAS with respect to homogeneous ones. An algorithm is developed for this task; building on a previous work on stability in distributed systems by Tad Hogg and Bernardo Huberman, and combining two phenomena observed in natural systems, task partition and hierarchical dominance. This algorithm is devised for allowing agents to learn which are the best tasks to perform on the basis of each agent's skills and the contribution to the team global performance. Agents learn by interacting with the environment and other teammates, and get rewards from the result of the actions they perform. This algorithm is specially designed for problems where all robots have to co-operate and work simultaneously towards the same goal. One example of such a problem is role distribution in a team of heterogeneous robots that form a soccer team, where all members take decisions and co-operate simultaneously. Soccer offers the possibility of conducting research in MAS, where co-operation plays a very important role in a dynamical and changing environment. For these reasons and the experience of the University of Girona in this domain, soccer has been selected as the test-bed for this research. In the case of soccer, tasks are grouped by means of roles. One of the most interesting features of this algorithm is that it endows MAS with a high adaptability to changes in the environment. It allows the team to perform their tasks, while adapting to the environment. This is studied in several cases, for changes in the environment and in the robot's body. Other features are also analysed, especially a parameter that defines the fitness (biological concept) of each agent in the system, which contributes to performance and team adaptability. The algorithm is applied later to allow agents to learn in teams of homogeneous and heterogeneous robots which roles they have to select, in order to maximise team performance. The teams are compared and the performance is evaluated in the games against three hand-coded teams and against the different homogeneous and heterogeneous teams built in this thesis. This section focuses on the analysis of performance and task partition, in order to study the benefits of heterogeneity in physical MAS. In order to study heterogeneity from a rigorous point of view, a diversity measure is developed building on the hierarchic social entropy defined by Tucker Balch. This is adapted to quantify physical diversity in robot teams. This tool presents very interesting features, as it can be used in the future to design heterogeneous teams on the basis of the knowledge on other teams.

Real-Time software control of neural stimulation in a closed loop Animat system

A recent area for investigation into the development of adaptable robot control is the use of living neuronal networks to control a mobile robot. The so-called Animat paradigm comprises a neuronal network (the ‘brain’) connected to an external embodiment (in this case a mobile robot), facilitating potentially robust, adaptable robot control and increased understanding of neural processes. Sensory input from the robot is provided to the neuronal network via stimulation on a number of electrodes embedded in a specialist Petri dish (Multi Electrode Array (MEA)); accurate control of this stimulation is vital. We present software tools allowing precise, near real-time control of electrical stimulation on MEAs, with fast switching between electrodes and the application of custom stimulus waveforms. These Linux-based tools are compatible with the widely used MEABench data acquisition system. Benefits include rapid stimulus modulation in response to neuronal activity (closed loop) and batch processing of stimulation protocols.

Organizational analysis of building projects

The principles of organization theory are applied to the organization of construction projects. This is done by proposing a framework for modelling the whole process of building procurement. This consists of a framework for describing the environments within which construction projects take place. This is followed by the development of a series of hypotheses about the organizational structure of construction projects. Four case studies are undertaken, and the extent to which their organizational structure matches the model is compared to the level of success achieved by each project. To this end there is a systematic method for evaluating the success of building project organizations, because any conclusions about the adequacy of a particular organization must be related to the degree of success achieved by that organization. In order to test these hypotheses, a mapping technique is developed. The technique offered is a development of a technique known as Linear Responsibility Analysis, and is called "3R analysis" as it deals with roles, responsibilities and relationships. The analysis of the case studies shows that they tended to suffer due to inappropriate organizational structure. One of the prevailing problems of public sector organization is that organizational structures are inadequately defined, and too cumbersome to respond to environmental demands on the project. The projects tended to be organized as rigid hierarchies, particularly at decision points, when what was required was a more flexible, dynamic and responsive organization. The study concludes with a series of recommendations; including suggestions for increasing the responsiveness of construction project organizations, and reducing the lead-in times for the inception periods.

Skill Rating by Bayesian Inference

Systems Engineering often involves computer modelling the behaviour of proposed systems and their components. Where a component is human, fallibility must be modelled by a stochastic agent. The identification of a model of decision-making over quantifiable options is investigated using the game-domain of Chess. Bayesian methods are used to infer the distribution of players’ skill levels from the moves they play rather than from their competitive results. The approach is used on large sets of games by players across a broad FIDE Elo range, and is in principle applicable to any scenario where high-value decisions are being made under pressure.

Review of theoretical models of electromagnetic NDT and their applications

There is an increasing interest in modelling electromagnetic methods of NDT - particularly eddy currents. A collaboration within the International Institute of Welding led to a survey intended to explain to non mathematicians the present scope of modelling. The present review commences with this survey and then points out some of the developments and some of the outstanding problems in transferring modelling into industry.

A likelihood ratio appropach to family-based association studies with covariates

We introduce a procedure for association based analysis of nuclear families that allows for dichotomous and more general measurements of phenotype and inclusion of covariate information. Standard generalized linear models are used to relate phenotype and its predictors. Our test procedure, based on the likelihood ratio, unifies the estimation of all parameters through the likelihood itself and yields maximum likelihood estimates of the genetic relative risk and interaction parameters. Our method has advantages in modelling the covariate and gene-covariate interaction terms over recently proposed conditional score tests that include covariate information via a two-stage modelling approach. We apply our method in a study of human systemic lupus erythematosus and the C-reactive protein that includes sex as a covariate.

A practical approach to a multi-level analysis with a sparse binary outcome within a large surgical trial

Objectives: To assess the potential source of variation that surgeon may add to patient outcome in a clinical trial of surgical procedures. Methods: Two large (n = 1380) parallel multicentre randomized surgical trials were undertaken to compare laparoscopically assisted hysterectomy with conventional methods of abdominal and vaginal hysterectomy; involving 43 surgeons. The primary end point of the trial was the occurrence of at least one major complication. Patients were nested within surgeons giving the data set a hierarchical structure. A total of 10% of patients had at least one major complication, that is, a sparse binary outcome variable. A linear mixed logistic regression model (with logit link function) was used to model the probability of a major complication, with surgeon fitted as a random effect. Models were fitted using the method of maximum likelihood in SAS((R)). Results: There were many convergence problems. These were resolved using a variety of approaches including; treating all effects as fixed for the initial model building; modelling the variance of a parameter on a logarithmic scale and centring of continuous covariates. The initial model building process indicated no significant 'type of operation' across surgeon interaction effect in either trial, the 'type of operation' term was highly significant in the abdominal trial, and the 'surgeon' term was not significant in either trial. Conclusions: The analysis did not find a surgeon effect but it is difficult to conclude that there was not a difference between surgeons. The statistical test may have lacked sufficient power, the variance estimates were small with large standard errors, indicating that the precision of the variance estimates may be questionable.

Comparison of mechanical properties of tension and opposite wood in Populus

In this paper we focused on the differences of mechanical properties of tension and normal wood of 1-year-old poplar trees, artificially tilted. Elastic and fracture properties have been measured and linked to the anatomy. Tension wood is well known because it prevents good surface finishing and leads to difficulties with sawing. We studied three main mechanical properties: young modulus, energy of cutting and longitudinal residual strain of maturation (with strain gauges) because of their importance in wood technology. Moreover, this work takes place in a larger project of study, the phenomena of axes re-orientation in trees (allowing by the production of reaction wood), where these data are required for biomechanical modelling. The results show that tension wood has a higher young modulus, needs a higher energy to be cut and exhibited a higher level of longitudinal residual strain of maturation than those of normal wood. The results suggest that these differences require deeper analysis of the wood than anatomy: measurement of microfibril orientation in the S2 layer and also the lignin composition in monomeric units.

Collaborative tele-rehabilitation: A strategy for increasing engagement

This paper examines the influence of a collaborative rehabilitation environment that encourages a long-distance collaborative "play" using two robot-mediated environments. This study presents a strategy for increasing motivation on able-bodied persons, applicable to impaired persons, to engage, sustain play and relate during a shared task. The study consisted of a series of eighteen single case studies, each involved in two distinct phases and assessed using a multidimensional measurement intended to assess participant' subjective experience. The results showed a clear positive trend in favour of the robot-mediated game environment. Subjects found the collaborative environment more valuable and more interesting and enjoyable. As a consequence, it appears subjects were willing to spend more time at a task.

Spectral characterisation of cooled filters for remote sensing

The improvements obtained on cooling atmospheric remote-sensing instruments for space flight applications has promoted research in characterization of the necessary optical filters. By modelling the effects of temperature on the dispersive spectrum of some constituent thin film materials, the cooled performance can be simulated and compared. multilayer filter designs with the measured spectra from actual filters. Two actual filters are discussed, for the 7µm region, one a composite cut-on/cut-off design of 13% HBW and the other an integral narrowband design of 4% HBW.

Minimum jerk trajectory control for rehabilitation and haptic applications

Smooth trajectories are essential for safe interaction in between human and a haptic interface. Different methods and strategies have been introduced to create such smooth trajectories. This paper studies the creation of human-like movements in haptic interfaces, based on the study of human arm motion. These motions are intended to retrain the upper limb movements of patients that lose manipulation functions following stroke. We present a model that uses higher degree polynomials to define a trajectory and control the robot arm to achieve minimum jerk movements. It also studies different methods that can be driven from polynomials to create more realistic human-like movements for therapeutic purposes.