Intelligent task-level grasp mapping for robot control

In the future, robots will enter our everyday lives to help us with various tasks.For a complete integration and cooperation with humans, these robots needto be able to acquire new skills. Sensor capabilities for navigation in real humanenvironments and intelligent interaction with humans are some of the keychallenges.Learning by demonstration systems focus on the problem of human robotinteraction, and let the human teach the robot by demonstrating the task usinghis own hands. In this thesis, we present a solution to a subproblem within thelearning by demonstration field, namely human-robot grasp mapping. Robotgrasping of objects in a home or office environment is challenging problem.Programming by demonstration systems, can give important skills for aidingthe robot in the grasping task.The thesis presents two techniques for human-robot grasp mapping, directrobot imitation from human demonstrator and intelligent grasp imitation. Inintelligent grasp mapping, the robot takes the size and shape of the object intoconsideration, while for direct mapping, only the pose of the human hand isavailable.These are evaluated in a simulated environment on several robot platforms.The results show that knowing the object shape and size for a grasping taskimproves the robot precision and performance

Towards Direct Policy Search Reinforcement Learning for Robot Control

This paper proposes a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot. Although the dominant approach, when using RL, has been to apply value function based algorithms, the system here detailed is characterized by the use of direct policy search methods. Rather than approximating a value function, these methodologies approximate a policy using an independent function approximator with its own parameters, trying to maximize the future expected reward. The policy based algorithm presented in this paper is used for learning the internal state/action mapping of a behavior. In this preliminary work, we demonstrate its feasibility with simulated experiments using the underwater robot GARBI in a target reaching task

On heuristic and linear models of judgment: Mapping the demand for knowledge

Research on judgment and decision making presents a confusing picture of human abilities. For example, much research has emphasized the dysfunctional aspects of judgmental heuristics, and yet, other findings suggest that these can be highly effective. A further line of research has modeled judgment as resulting from as if linear models. This paper illuminates the distinctions in these approaches by providing a common analytical framework based on the central theoretical premise that understanding human performance requires specifying how characteristics of the decision rules people use interact with the demands of the tasks they face. Our work synthesizes the analytical tools of lens model research with novel methodology developed to specify the effectiveness of heuristics in different environments and allows direct comparisons between the different approaches. We illustrate with both theoretical analyses and simulations. We further link our results to the empirical literature by a meta-analysis of lens model studies and estimate both human andheuristic performance in the same tasks. Our results highlight the trade-off betweenlinear models and heuristics. Whereas the former are cognitively demanding, the latterare simple to use. However, they require knowledge and thus maps of when andwhich heuristic to employ.

Evidence on the determinants of foreign direct investment: the case of three european regions

This study aims at analyzing the determinants of FDI (foreign direct investment) inflows for a group of European regions. The originality of this approach lies in the use of disaggregated regional data. First, we develop a qualitative description of our database and discuss the importance of the macroeconomic determinants in attracting FDI. Then, we provide an econometric exercise to identify the potential determinants of FDI. In spite of choosing regions presenting economic similarities, we show that regional FDI inflows rely on a combination of factors that differs from one region to another.

Do creative industries cluster? mapping creative local production systems in Italy

An important debate on the role of creativity and culture as factors of local economic development is distinctly emerging. Despite the emphasis put on the theoretical definition of these concepts, it is necessary to strengthen comparative research for the identification and analysis of the kind of creativity embedded in the territory as well as its determinants. Creative local production systems are identified in Italy and Spain departing from local labour markets as territorial units, and focusing on two different kinds of creative

Association mapping of candidate genes for drought tolerance in Mediterranean durum wheat landraces

Treball al que se li ha concedit el premi al millor póster del congrés

A first attempt of geographically-distributed Multi-scale integrated analysis of societal and ecosystem metabolism (MuSIASEM): mapping human time and energy throughput in metropolitan Barcelona

This study presents a first attempt to extend the “Multi-scale integrated analysis of societal and ecosystem metabolism (MuSIASEM)” approach to a spatial dimension using GIS techniques in the Metropolitan area of Barcelona. We use a combination of census and commercial databases along with a detailed land cover map to create a layer of Common Geographic Units that we populate with the local values of human time spent in different activities according to MuSIASEM hierarchical typology. In this way, we mapped the hours of available human time, in regards to the working hours spent in different locations, putting in evidence the gradients in spatial density between the residential location of workers (generating the work supply) and the places where the working hours are actually taking place. We found a strong three-modal pattern of clumps of areas with different combinations of values of time spent on household activities and on paid work. We also measured and mapped spatial segregation between these two activities and put forward the conjecture that this segregation increases with higher energy throughput, as the size of the functional units must be able to cope with the flow of exosomatic energy. Finally, we discuss the effectiveness of the approach by comparing our geographic representation of exosomatic throughput to the one issued from conventional methods.

MR MAQ: algorisme de Read Mapping utilitzant la plataforma Hadoop

L’èxit del Projecte Genoma Humà (PGH) l’any 2000 va fer de la “medicina personalitzada” una realitat més propera. Els descobriments del PGH han simplificat les tècniques de seqüenciació de tal manera que actualment qualsevol persona pot aconseguir la seva seqüència d’ADN complerta. La tecnologia de Read Mapping destaca en aquest tipus de tècniques i es caracteritza per manegar una gran quantitat de dades. Hadoop, el framework d’Apache per aplicacions intensives de dades sota el paradigma Map Reduce, resulta un aliat perfecte per aquest tipus de tecnologia i ha sigut l’opció escollida per a realitzar aquest projecte. Durant tot el treball es realitza l’estudi, l’anàlisi i les experimentacions necessàries per aconseguir un Algorisme Genètic innovador que utilitzi tot el potencial de Hadoop.

The nature of the Cr-C bond: direct estimation of aromaticity in Fischer carbenes

Report for the scientific sojourn at the the Philipps-Universität Marburg, Germany, from september to december 2007. For the first, we employed the Energy-Decomposition Analysis (EDA) to investigate aromaticity on Fischer carbenes as it is related through all the reaction mechanisms studied in my PhD thesis. This powerful tool, compared with other well-known aromaticity indices in the literature like NICS, is useful not only for quantitative results but also to measure the degree of conjugation or hyperconjugation in molecules. Our results showed for the annelated benzenoid systems studied here, that electron density is more concentrated on the outer rings than in the central one. The strain-induced bond localization plays a major role as a driven force to keep the more substituted ring as the less aromatic. The discussion presented in this work was contrasted at different levels of theory to calibrate the method and ensure the consistency of our results. We think these conclusions can also be extended to arene chemistry for explaining aromaticity and regioselectivity reactions found in those systems.In the second work, we have employed the Turbomole program package and density-functionals of the best performance in the state of art, to explore reaction mechanisms in the noble gas chemistry. Particularly, we were interested in compounds of the form H--Ng--Ng--F (where Ng (Noble Gas) = Ar, Kr and Xe) and we investigated the relative stability of these species. Our quantum chemical calculations predict that the dixenon compound HXeXeF has an activation barrier for decomposition of 11 kcal/mol which should be large enough to identify the molecule in a low-temperature matrix. The other noble gases present lower activation barriers and therefore are more labile and difficult to be observable systems experimentally.

Straightening out the concept of direct and indirect input requirements

Gim & Kim (1998) proposed a generalization of Jeong (1982, 1984) reinterpretation of the Hawkins-Simon condition for macroeconomic stability to off-diagonal matrix elements. This generalization is conceptually relevant for it offers a complementary view of interindustry linkages beyond final or net output influence. The extension is completely similar to the 'total flow' idea introduced by Szyrmer (1992) or the 'output-to-output' multiplier of Miller & Blair (2009). However the practical implementation of Gim & Kim is actually faulty since it confuses the appropriate order of output normalization. We provide a new and elementary solution for the correct formalization using standard interindustry accounting concepts.

Cortical mapping of the neuronal circuits modulating the muscle tone. Introduction to the electrophysiological treatment of the spastic hand.

L’objectiu d’aquest estudi es investigar l’organització cortical junt amb la connectivitat còrtico-subcortical en subjectes sans, com a estudi preliminar. Els mapes corticals s’han fet per TMS navegada, i els punts motors obtinguts s’han exportant per estudi tractogràfic i anàlisi de las seves connexions. El coneixement precís de la localització de l’àrea cortical motora primària i les seves connexions es la base per ser utilitzada en estudis posteriors de la reorganització cortical i sub-cortical en pacients amb infart cerebral. Aquesta reorganització es deguda a la neuroplasticitat i pot ser influenciada per els efectes neuromoduladors de la estimulació cerebral no invasiva.

Policy gradient based Reinforcement Learning for real autonomous underwater cable tracking

This paper proposes a field application of a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot in cable tracking task. The learning system is characterized by using a direct policy search method for learning the internal state/action mapping. Policy only algorithms may suffer from long convergence times when dealing with real robotics. In order to speed up the process, the learning phase has been carried out in a simulated environment and, in a second step, the policy has been transferred and tested successfully on a real robot. Future steps plan to continue the learning process on-line while on the real robot while performing the mentioned task. We demonstrate its feasibility with real experiments on the underwater robot ICTINEU AUV

Autonomous underwater vehicle control using reinforcement learning policy search methods

Autonomous underwater vehicles (AUV) represent a challenging control problem with complex, noisy, dynamics. Nowadays, not only the continuous scientific advances in underwater robotics but the increasing number of subsea missions and its complexity ask for an automatization of submarine processes. This paper proposes a high-level control system for solving the action selection problem of an autonomous robot. The system is characterized by the use of reinforcement learning direct policy search methods (RLDPS) for learning the internal state/action mapping of some behaviors. We demonstrate its feasibility with simulated experiments using the model of our underwater robot URIS in a target following task

Online Robust 3D Mapping Using Structure from Motion Cues

This paper presents a complete solution for creating accurate 3D textured models from monocular video sequences. The methods are developed within the framework of sequential structure from motion, where a 3D model of the environment is maintained and updated as new visual information becomes available. The camera position is recovered by directly associating the 3D scene model with local image observations. Compared to standard structure from motion techniques, this approach decreases the error accumulation while increasing the robustness to scene occlusions and feature association failures. The obtained 3D information is used to generate high quality, composite visual maps of the scene (mosaics). The visual maps are used to create texture-mapped, realistic views of the scene

Mapping sub-flows to p2mp LSPs

In previous work we proposed a multi-objective traffic engineering scheme (MHDB-S model) using different distribution trees to multicast several flows. In this paper, we propose a heuristic algorithm to create multiple point-to-multipoint (p2mp) LSPs based on the optimum sub-flow values obtained with our MHDB-S model. Moreover, a general problem for supporting multicasting in MPLS networks is the lack of labels. To reduce the number of labels used, a label space reduction algorithm solution is also considered