http://hdl.handle.net/2117/2252 Sat, 25 May 2013 06:11:57 GMT 2013-05-25T06:11:57Z webmaster.bupc@upc.edu Universitat Politècnica de Catalunya. Servei de Biblioteques i Documentació no http://hdl.handle.net/2117/15641 Title: Branch switching from singular points in higher-dimensional continuation Authors: Bohigas Nadal, Oriol Abstract: We explain here how to perform branch switching when a singular point is found during higherdimensional continuation on a k-dimensional variety. This document is based on the information given in [1, 2, 3]. Wed, 21 Mar 2012 19:12:23 GMT http://hdl.handle.net/2117/15641 2012-03-21T19:12:23Z Bohigas Nadal, Oriol no robot kinematics PARAULES AUTOR: higher-dimensional continuation, bifurcation, branch switching, manifold, differentiable variety, singularity We explain here how to perform branch switching when a singular point is found during higherdimensional continuation on a k-dimensional variety. This document is based on the information given in [1, 2, 3]. http://hdl.handle.net/2117/14112 Title: Stochastic approximations of average values using proportions of samples Authors: Agostini, Alejandro Gabriel; Celaya Llover, Enric Abstract: In this work we explain how the stochastic approximation of the average of a random variable is carried out when the observations used in the updates consist in proportion of samples rather than complete samples. Description: IRI Technical Report Tue, 29 Nov 2011 14:54:00 GMT http://hdl.handle.net/2117/14112 2011-11-29T14:54:00Z Agostini, Alejandro Gabriel; Celaya Llover, Enric no In this work we explain how the stochastic approximation of the average of a random variable is carried out when the observations used in the updates consist in proportion of samples rather than complete samples. http://hdl.handle.net/2117/13951 Title: A general strategy for interactive decision-making in robotic platforms Authors: Agostini, Alejandro Gabriel; Torras, Carme; Wörgötter, Florentin Abstract: This work presents an intergated strategy for planning and learning suitable to execute tasks with robotic platforms without any previous task specification. The approach rapidly learns planning operators from few action experiences using a competitive strategy where many alternatives of cause-effect explanations are evaluated in parallel, and the most successful ones are used to generate the operators. The system operates without task interruption by integrating in the planning-learning loop a human teacher that supports the planner in making decisions. All the mechanisms are integrated and synchronized in the robot using a general decision-making framework. Thu, 17 Nov 2011 10:49:57 GMT http://hdl.handle.net/2117/13951 2011-11-17T10:49:57Z Agostini, Alejandro Gabriel; Torras, Carme; Wörgötter, Florentin no Learning (artificial intelligence) Planning (artificial intelligence) This work presents an intergated strategy for planning and learning suitable to execute tasks with robotic platforms without any previous task specification. The approach rapidly learns planning operators from few action experiences using a competitive strategy where many alternatives of cause-effect explanations are evaluated in parallel, and the most successful ones are used to generate the operators. The system operates without task interruption by integrating in the planning-learning loop a human teacher that supports the planner in making decisions. All the mechanisms are integrated and synchronized in the robot using a general decision-making framework. http://hdl.handle.net/2117/13936 Title: Registration of 3d point clouds for urban robot mapping Authors: Teniente Avilés, Ernesto; Andrade-Cetto, Juan Abstract: We consider the task of mapping pedestrian urban areas for a robotic guidance and surveillance application. This mapping is performed by registering three-dimensional laser range scans acquired with two different robots. To solve this task we will use the Iterative Closes Point (ICP) algorithm proposed in [8], but for the minimization step we will use the metric proposed by Biota et al. [10] trying to get advantage of the compensation between translation and rotation they mention. To reduce computational cost in the original ICP during matching, the correspondences search is done with the library Approximate Nearest Neighbor (ANN). Finally we propose a hierarchical new correspondence search strategy, using a point-to-plane strategy at the highest level and the point-to-point metric at finer levels. At the highest level the adjust error between a plane and it’s n adjacent points describing the plane is computed, if this error is bigger than a threshold then we change the level. Wed, 16 Nov 2011 13:18:24 GMT http://hdl.handle.net/2117/13936 2011-11-16T13:18:24Z Teniente Avilés, Ernesto; Andrade-Cetto, Juan no Automation 3d mapping 3d registration ICP We consider the task of mapping pedestrian urban areas for a robotic guidance and surveillance application. This mapping is performed by registering three-dimensional laser range scans acquired with two different robots. To solve this task we will use the Iterative Closes Point (ICP) algorithm proposed in [8], but for the minimization step we will use the metric proposed by Biota et al. [10] trying to get advantage of the compensation between translation and rotation they mention. To reduce computational cost in the original ICP during matching, the correspondences search is done with the library Approximate Nearest Neighbor (ANN). Finally we propose a hierarchical new correspondence search strategy, using a point-to-plane strategy at the highest level and the point-to-point metric at finer levels. At the highest level the adjust error between a plane and it’s n adjacent points describing the plane is computed, if this error is bigger than a threshold then we change the level. http://hdl.handle.net/2117/12449 Title: Path planning with pose SLAM Authors: Valencia Carreño, Rafael; Andrade-Cetto, Juan; Porta Pleite, Josep Maria Abstract: The probabilistic belief networks that result from standard feature-based simultaneous localization and map building (SLAM) approaches cannot be directly used to plan trajectories. The reason is that they produce a sparse graph of landmark estimates and their probabilistic relations, which is of little value to find collision free paths for navigation. In contrast, we argue in this paper that Pose SLAM graphs can be directly used as belief roadmaps (BRMs). The original BRM algorithm assumes a known model of the environment from which probabilistic sampling generates a roadmap. In our work, the roadmap is built on-line by the Pose SLAM algorithm. The result is a hybrid BRM-Pose SLAM method that devises optimal navigation strategies on-line by searching for the path with lowest accumulated uncertainty for the robot pose. The method is validated over synthetic data and standard SLAM datasets. Tue, 03 May 2011 09:44:15 GMT http://hdl.handle.net/2117/12449 2011-05-03T09:44:15Z Valencia Carreño, Rafael; Andrade-Cetto, Juan; Porta Pleite, Josep Maria no Path Planning, SLAM The probabilistic belief networks that result from standard feature-based simultaneous localization and map building (SLAM) approaches cannot be directly used to plan trajectories. The reason is that they produce a sparse graph of landmark estimates and their probabilistic relations, which is of little value to find collision free paths for navigation. In contrast, we argue in this paper that Pose SLAM graphs can be directly used as belief roadmaps (BRMs). The original BRM algorithm assumes a known model of the environment from which probabilistic sampling generates a roadmap. In our work, the roadmap is built on-line by the Pose SLAM algorithm. The result is a hybrid BRM-Pose SLAM method that devises optimal navigation strategies on-line by searching for the path with lowest accumulated uncertainty for the robot pose. The method is validated over synthetic data and standard SLAM datasets. http://hdl.handle.net/2117/12364 Title: Quick learning of cause-effects relevant for robot action Authors: Agostini, Alejandro Gabriel; Wörgötter, Florentin; Torras, Carme Abstract: In this work we propose a new paradigm for the rapid learning of cause-effect relations relevant for task execution. Learning occurs automatically from action experiences by means of a novel constructive learning approach designed for applications where there is no previous knowledge of the task or world model, examples are provided on-line during run time, and the number of examples is small compared to the number of incoming experiences. These limitations pose obstacles for the existing constructive learning methods, where on-line learning is either not considered, a significant amount of prior knowledge has to be provided, or a large number of experiences or training streams are required. The system is implemented and evaluated in a humanoid robot platform using a decision-making framework that integrates a planner, the proposed learning mechanism, and a human teacher that supports the planner in the action selection. Results demonstrate the feasibility of the system for decision making in robotic applications. Wed, 13 Apr 2011 17:03:49 GMT http://hdl.handle.net/2117/12364 2011-04-13T17:03:49Z Agostini, Alejandro Gabriel; Wörgötter, Florentin; Torras, Carme no learning (artificial intelligence) service robots. In this work we propose a new paradigm for the rapid learning of cause-effect relations relevant for task execution. Learning occurs automatically from action experiences by means of a novel constructive learning approach designed for applications where there is no previous knowledge of the task or world model, examples are provided on-line during run time, and the number of examples is small compared to the number of incoming experiences. These limitations pose obstacles for the existing constructive learning methods, where on-line learning is either not considered, a significant amount of prior knowledge has to be provided, or a large number of experiences or training streams are required. The system is implemented and evaluated in a humanoid robot platform using a decision-making framework that integrates a planner, the proposed learning mechanism, and a human teacher that supports the planner in the action selection. Results demonstrate the feasibility of the system for decision making in robotic applications. http://hdl.handle.net/2117/8685 Title: Diseño de un pie para un robot humanoide Authors: Barbadillo Villanueva, Guillermo; Alenyà Ribas, Guillem Abstract: La propuesta se enmarca dentro del proyecto Humanoid Lab del Institut de Robòtica i Informàtica Industrial (IRI). El grupo dispone de múltiples plataformas humanoides educativas (Robonova y Bioloid). Existe una primera versión de un sistema de realimentación de fuerza para los pies de estos robots, desarrollada en el propio grupo, que funciona correctamente pero presenta algunas deficiencias y limitaciones que se pretenden subsanar. El objetivo de este trabajo es diseñar e implementar un nuevo sistema de sensores del pie. Mon, 23 Aug 2010 09:27:33 GMT http://hdl.handle.net/2117/8685 2010-08-23T09:27:33Z Barbadillo Villanueva, Guillermo; Alenyà Ribas, Guillem no La propuesta se enmarca dentro del proyecto Humanoid Lab del Institut de Robòtica i Informàtica Industrial (IRI). El grupo dispone de múltiples plataformas humanoides educativas (Robonova y Bioloid). Existe una primera versión de un sistema de realimentación de fuerza para los pies de estos robots, desarrollada en el propio grupo, que funciona correctamente pero presenta algunas deficiencias y limitaciones que se pretenden subsanar. El objetivo de este trabajo es diseñar e implementar un nuevo sistema de sensores del pie. http://hdl.handle.net/2117/8223 Title: Exploitation of time-of-flight (ToF) cameras Authors: Foix Salmerón, Sergi; Alenyà Ribas, Guillem; Torras, Carme Abstract: This technical report reviews the state-of-the art in the field of ToF cameras, their advantages, their limitations, and their present-day applications sometimes in combination with other sensors. Even though ToF cameras provide neither higher resolution nor larger ambiguity-free range compared to other range map estimation systems, advantages such as registered depth and intensity data at a high frame rate, compact design, low weight and reduced power consumption have motivated their use in numerous areas of research. In robotics, these areas range from mobile robot navigation and map building to vision-based human motion capture and gesture recognition, showing particularly a great potential in object modeling and recognition. Mon, 19 Jul 2010 08:00:33 GMT http://hdl.handle.net/2117/8223 2010-07-19T08:00:33Z Foix Salmerón, Sergi; Alenyà Ribas, Guillem; Torras, Carme no Time-of-Flight ToF cameras Review This technical report reviews the state-of-the art in the field of ToF cameras, their advantages, their limitations, and their present-day applications sometimes in combination with other sensors. Even though ToF cameras provide neither higher resolution nor larger ambiguity-free range compared to other range map estimation systems, advantages such as registered depth and intensity data at a high frame rate, compact design, low weight and reduced power consumption have motivated their use in numerous areas of research. In robotics, these areas range from mobile robot navigation and map building to vision-based human motion capture and gesture recognition, showing particularly a great potential in object modeling and recognition. http://hdl.handle.net/2117/7502 Title: Multi-body singularity equations Authors: Bohigas Nadal, Oriol; Ros Giralt, Lluís Abstract: This technical report explains how to obtain a system of equations that encodes the singularities of a multi-body system with respect to some of its configuration variables. The system is obtained in a way that makes it appropiate for the CUIK software. Thu, 03 Jun 2010 12:38:17 GMT http://hdl.handle.net/2117/7502 2010-06-03T12:38:17Z Bohigas Nadal, Oriol; Ros Giralt, Lluís no Singularities Multi-body CUIK This technical report explains how to obtain a system of equations that encodes the singularities of a multi-body system with respect to some of its configuration variables. The system is obtained in a way that makes it appropiate for the CUIK software. http://hdl.handle.net/2117/6856 Title: Probability density estimation of the Q Function for reinforcement learning Authors: Agostini, Alejandro Gabriel; Celaya Llover, Enric Abstract: Performing Q-Learning in continuous state-action spaces is a problem still unsolved for many complex applications. The Q function may be rather complex and can not be expected to fit into a predefined parametric model. In addition, the function approximation must be able to cope with the high non-stationarity of the estimated q values, the on-line nature of the learning with a strongly biased sampling to convergence regions, and the large amount of generalization required for a feasible implementation. To cope with these problems local, non-parametric function approximations seem more suitable than global parametric ones. A kind of function approximation that is gaining special interest in the field of machine learning are those based on densities. Estimating densities provides more information than simple function approximations which can be used to deal with the Reinforcement Learning problems. For instance, density estimation permits to know the actual distribution of the q values for any given state-action, and provides information about how many data has been collected in different regions of the domain. In this work we propose a Q-Learning approach for continuous state-action spaces based on joint density estimations. The density distribution is represented with a Gaussian Mixture Model using an on-line version of the Expectation-Maximization algorithm. We propose a method that handles the biased sampling problem with good performance. Experiments performed on a test problem show remarkable improvements over previous published results. Thu, 01 Apr 2010 11:30:53 GMT http://hdl.handle.net/2117/6856 2010-04-01T11:30:53Z Agostini, Alejandro Gabriel; Celaya Llover, Enric no Performing Q-Learning in continuous state-action spaces is a problem still unsolved for many complex applications. The Q function may be rather complex and can not be expected to fit into a predefined parametric model. In addition, the function approximation must be able to cope with the high non-stationarity of the estimated q values, the on-line nature of the learning with a strongly biased sampling to convergence regions, and the large amount of generalization required for a feasible implementation. To cope with these problems local, non-parametric function approximations seem more suitable than global parametric ones. A kind of function approximation that is gaining special interest in the field of machine learning are those based on densities. Estimating densities provides more information than simple function approximations which can be used to deal with the Reinforcement Learning problems. For instance, density estimation permits to know the actual distribution of the q values for any given state-action, and provides information about how many data has been collected in different regions of the domain. In this work we propose a Q-Learning approach for continuous state-action spaces based on joint density estimations. The density distribution is represented with a Gaussian Mixture Model using an on-line version of the Expectation-Maximization algorithm. We propose a method that handles the biased sampling problem with good performance. Experiments performed on a test problem show remarkable improvements over previous published results. http://hdl.handle.net/2117/6840 Title: Robot learning of container-emptying skills through haptic demonstration Authors: Rozo Castañeda, Leonel; Jimenez Schlegl, Pablo; Torras, Carme Abstract: Locally weighted learning algorithms are suitable strategies for trajectory learning and skill acquisition, in the context of programming by demonstration. Input streams other than visual information, as used in most applications up to date, reveal themselves as quite useful in trajectory learning experiments where visual sources are not available. In this work we have used force/torque feedback through a haptic device for teaching a teleoperated robot to empty a rigid container. Structure vibrations and container inertia appeared to considerably disrupt the sensing process, so a filtering algorithm had to be devised. Then, the memory-based LWPLS and the non-memory-based LWPR algorithms [8, 13, 10] were implemented, their comparison leading to very similar results, with the same pattern as regards to both the involved robot joints and the different initial experimental conditions. Tests where the teacher was instructed to follow a strategy compared to others where he was not lead to useful conclusions that permit devising the new research stages, where the taught motion will be refined by autonomous robot rehearsal through reinforcement learning. Tue, 30 Mar 2010 11:39:49 GMT http://hdl.handle.net/2117/6840 2010-03-30T11:39:49Z Rozo Castañeda, Leonel; Jimenez Schlegl, Pablo; Torras, Carme no Locally weighted learning algorithms are suitable strategies for trajectory learning and skill acquisition, in the context of programming by demonstration. Input streams other than visual information, as used in most applications up to date, reveal themselves as quite useful in trajectory learning experiments where visual sources are not available. In this work we have used force/torque feedback through a haptic device for teaching a teleoperated robot to empty a rigid container. Structure vibrations and container inertia appeared to considerably disrupt the sensing process, so a filtering algorithm had to be devised. Then, the memory-based LWPLS and the non-memory-based LWPR algorithms [8, 13, 10] were implemented, their comparison leading to very similar results, with the same pattern as regards to both the involved robot joints and the different initial experimental conditions. Tests where the teacher was instructed to follow a strategy compared to others where he was not lead to useful conclusions that permit devising the new research stages, where the taught motion will be refined by autonomous robot rehearsal through reinforcement learning. http://hdl.handle.net/2117/6434 Title: Reducció de variables en problemes de control predictiu amb restriccions d'igualtat Authors: Joseph Duran, Bernat; Ocampo-Martínez, Carlos Description: En el següent document es mostra una técnica per utilitzar les restriccions d'igualtat associades a un problema de control predictiu per tal de reduir el nombre de variables que intervenen en aquest problema. Primerament fem un repàs de les notacions i desenvolupament del problema estàndard per veure, després, que mitjançant un canvi lineal el nou problema, amb un nombre inferior de variables, té una estructura completament anàloga. Alguns resultats de temps de computació mostren, en la última secció, la utilitat del mètode. Mon, 22 Feb 2010 13:12:51 GMT http://hdl.handle.net/2117/6434 2010-02-22T13:12:51Z Joseph Duran, Bernat; Ocampo-Martínez, Carlos no http://hdl.handle.net/2117/2698 Title: Exploiting single-cycle symmetries in continuous constraint problems Authors: Ruiz de Angulo García, Vicente; Torras, Carme Abstract: Symmetries in discrete constraint satisfaction problems have been explored and exploited in the last years, but symmetries in continuous constraint problems have not received the same attention. Here we focus on permutations of the variables consisting of one single cycle. We propose a procedure that takes advantage of these symmetries by interacting with a continuous constraint solver without interfering with it. A key concept in this procedure are the classes of symmetric boxes formed by bisecting a n-dimensional cube at the same point in all dimensions at the same time. We analyze these classes and quantify them as a function of the cube dimensionality. Moreover, we propose a simple algorithm to generate the representatives of all these classes for any number of variables at very high rates. A problem example from the chemical field and a kinematics solver are used to show the performance of the approach in practice. Fri, 13 Mar 2009 09:24:40 GMT http://hdl.handle.net/2117/2698 2009-03-13T09:24:40Z Ruiz de Angulo García, Vicente; Torras, Carme no symmetries, continuous constraint problems Symmetries in discrete constraint satisfaction problems have been explored and exploited in the last years, but symmetries in continuous constraint problems have not received the same attention. Here we focus on permutations of the variables consisting of one single cycle. We propose a procedure that takes advantage of these symmetries by interacting with a continuous constraint solver without interfering with it. A key concept in this procedure are the classes of symmetric boxes formed by bisecting a n-dimensional cube at the same point in all dimensions at the same time. We analyze these classes and quantify them as a function of the cube dimensionality. Moreover, we propose a simple algorithm to generate the representatives of all these classes for any number of variables at very high rates. A problem example from the chemical field and a kinematics solver are used to show the performance of the approach in practice. http://hdl.handle.net/2117/2696 Title: Map format for mobile robot map-based autonomous navigation Authors: Corominas Murtra, Andreu; Mirats Tur, Josep Maria Abstract: This technical report defines the spatial representation and the map file format used in a mobile robot map-based autonomous navigation system designed to be deployed in urban areas. After a discussion about common requirements of spatial representations for map-based mobile robot autonomous navigation, a proposed environment model that fulfills previously discussed requirements is formally presented. An example of a map representing an outdoor area of an university campus of about 10000m2 is given to better illustrate the map format. Finally, the report shows simulation results on global localization and path planning using the proposed map. Fri, 13 Mar 2009 09:24:26 GMT http://hdl.handle.net/2117/2696 2009-03-13T09:24:26Z Corominas Murtra, Andreu; Mirats Tur, Josep Maria no This technical report defines the spatial representation and the map file format used in a mobile robot map-based autonomous navigation system designed to be deployed in urban areas. After a discussion about common requirements of spatial representations for map-based mobile robot autonomous navigation, a proposed environment model that fulfills previously discussed requirements is formally presented. An example of a map representing an outdoor area of an university campus of about 10000m2 is given to better illustrate the map format. Finally, the report shows simulation results on global localization and path planning using the proposed map. http://hdl.handle.net/2117/2694 Title: On-line learning of macro planning operators using probabilistic estimations of cause-effects Authors: Agostini, Alejandro Gabriel; Wörgötter, Florentin; Celaya Llover, Enric; Torras, Carme Abstract: In this work we propose an on-line learning method for learning action rules for planning. The system uses a probabilistic approach of a constructive induction method that combines a beam search with an example-based search over candidate rules to find those that more concisely describe the world dynamics. The approach permits a rapid integration of the knowledge acquired from experience. Exploration of the world dynamics is guided by the planner, and – if the planner fails because of incomplete knowledge – by a teacher through action instructions. Fri, 13 Mar 2009 09:24:12 GMT http://hdl.handle.net/2117/2694 2009-03-13T09:24:12Z Agostini, Alejandro Gabriel; Wörgötter, Florentin; Celaya Llover, Enric; Torras, Carme no online learning, macro planning operator, constructive learning In this work we propose an on-line learning method for learning action rules for planning. The system uses a probabilistic approach of a constructive induction method that combines a beam search with an example-based search over candidate rules to find those that more concisely describe the world dynamics. The approach permits a rapid integration of the knowledge acquired from experience. Exploration of the world dynamics is guided by the planner, and – if the planner fails because of incomplete knowledge – by a teacher through action instructions.