Learning Robotics for youngsters - the RoboParty experience

Series: "Advances in intelligent systems and computing , ISSN 2194-5357, vol. 417"

Techniques for batch reinforcement learning in robotics

This thesis addresses the Batch Reinforcement Learning methods in Robotics. This sub-class of Reinforcement Learning has shown promising results and has been the focus of recent research. Three contributions are proposed that aim to extend the state-of-art methods allowing for a faster and more stable learning process, such as required for learning in Robotics. The Q-learning update-rule is widely applied, since it allows to learn without the presence of a model of the environment. However, this update-rule is transition-based and does not take advantage of the underlying episodic structure of collected batch of interactions. The Q-Batch update-rule is proposed in this thesis, to process experiencies along the trajectories collected in the interaction phase. This allows a faster propagation of obtained rewards and penalties, resulting in faster and more robust learning. Non-parametric function approximations are explored, such as Gaussian Processes. This type of approximators allows to encode prior knowledge about the latent function, in the form of kernels, providing a higher level of exibility and accuracy. The application of Gaussian Processes in Batch Reinforcement Learning presented a higher performance in learning tasks than other function approximations used in the literature. Lastly, in order to extract more information from the experiences collected by the agent, model-learning techniques are incorporated to learn the system dynamics. In this way, it is possible to augment the set of collected experiences with experiences generated through planning using the learned models. Experiments were carried out mainly in simulation, with some tests carried out in a physical robotic platform. The obtained results show that the proposed approaches are able to outperform the classical Fitted Q Iteration.

A learning environment for robotic production line simulation

This paper presents a framework for a robotic production line simulation learning environment using Autonomous Ground Vehicles (AGV). An eLearning platform is used as interface with the simulator. The objective is to introduce students to the production robotics area using a familiar tool, an eLearning platform, and a framework that simulates a production line using AGVs. This framework allows students to learn about robotics but also about several areas of industrial management engineering without requiring an extensive prior knowledge on the robotics area. The robotic production line simulation learning environment simulates a production environment using AGVs to transport materials to and from the production line. The simulator allows students to validate the AGV dynamics and provides information about the whole materials supplying system which includes: supply times, route optimization and inventory management. The students are required to address several topics such as: sensors, actuators, controllers and an high level management and optimization software. This simulator was developed with a known open source tool from robotics community: Player/Stage. This tool was extended with several add-ons so that students can be able to interact with a complex simulation environment. These add-ons include an abstraction communication layer that performs events provided by the database server which is programmed by the students. An eLearning platform is used as interface between the students and the simulator. The students can visualize the effects of their instructions/programming in the simulator that they can access via the eLearning platform. The proposed framework aims to allow students from different backgrounds to fully experience robotics in practice by suppressing the huge gap between theory and practice that exists in robotics. Using an eLearning platform eliminates installation problems that can occur from different computers software distribution and makes the simulator accessible by all students at school and at home.

Structure learning of graphical models for task-oriented robot grasping

In the collective imaginaries a robot is a human like machine as any androids in science fiction. However the type of robots that you will encounter most frequently are machinery that do work that is too dangerous, boring or onerous. Most of the robots in the world are of this type. They can be found in auto, medical, manufacturing and space industries. Therefore a robot is a system that contains sensors, control systems, manipulators, power supplies and software all working together to perform a task. The development and use of such a system is an active area of research and one of the main problems is the development of interaction skills with the surrounding environment, which include the ability to grasp objects. To perform this task the robot needs to sense the environment and acquire the object informations, physical attributes that may influence a grasp. Humans can solve this grasping problem easily due to their past experiences, that is why many researchers are approaching it from a machine learning perspective finding grasp of an object using information of already known objects. But humans can select the best grasp amongst a vast repertoire not only considering the physical attributes of the object to grasp but even to obtain a certain effect. This is why in our case the study in the area of robot manipulation is focused on grasping and integrating symbolic tasks with data gained through sensors. The learning model is based on Bayesian Network to encode the statistical dependencies between the data collected by the sensors and the symbolic task. This data representation has several advantages. It allows to take into account the uncertainty of the real world, allowing to deal with sensor noise, encodes notion of causality and provides an unified network for learning. Since the network is actually implemented and based on the human expert knowledge, it is very interesting to implement an automated method to learn the structure as in the future more tasks and object features can be introduced and a complex network design based only on human expert knowledge can become unreliable. Since structure learning algorithms presents some weaknesses, the goal of this thesis is to analyze real data used in the network modeled by the human expert, implement a feasible structure learning approach and compare the results with the network designed by the expert in order to possibly enhance it.

Desenvolvimento do sistema de controlo de um braço robotizado e respetiva interface com o utilizador

A robótica tem evoluído de forma significativa nos últimos anos e passa a ser indispensável em várias aplicações nas áreas da engenharia, aeronáutica, medicina, entre outras. O estado da arte do presente trabalho está dividido em duas partes, uma que aborda vários aspetos relacionados com a robótica e outra com os aspetos da fundamentação matemática por de trás da robótica, porque para controlar o robô é necessário implementar expressões matemáticas para o poder controlar. Neste trabalho é apresentado um sistema de controlo do braço robótico MENTOR e o desenvolvimento de uma interface para o utilizador. Para o controlo do braço robótico foi necessário calcular a cinemática direta e inversa, para que se possa obter os ângulos das juntas para uma dada posição ou qual é a posição final do braço robótico para um valor das juntas. O sistema é bastante flexível e foi desenvolvido para ser utilizado essencialmente para aprendizagem de robótica, podendo no entanto ser utilizado em outras aplicações.

Recursos y herramientas didácticas para el aprendizaje de la robótica

Este artículo analiza diferentes experiencias docentes que tienen como finalidad el aprendizaje de la robótica en el mundo universitario. Estas experiencias se plasman en el desarrollo de varios cursos y asignaturas sobre robótica que se imparten en la Universidad de Alicante. Para el desarrollo de estos cursos, los autores han empleado varias plataformas educativas, algunas de implementación propia, otras de libre distribución y código abierto. El objetivo de estos cursos es enseñar el diseño e implementación de soluciones robóticas a diversos problemas que van desde el control, programación y manipulación de brazos robots de ámbito industrial hasta la construcción y/o programación de mini-robots con carácter educativo. Por un lado, se emplean herramientas didácticas de última generación como simuladores y laboratorios virtuales que flexibilizan el uso de brazos robots y, por otro lado, se hace uso de competiciones y concursos para motivar al alumno haciendo que ponga en práctica las destrezas aprendidas, mediante la construcción y programación de mini-robots de bajo coste.

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.

Learning of shared attention in sociable robotics

Sociable robots are embodied agents that are part of a heterogeneous society of robots and humans. They Should be able to recognize human beings and each other, and to engage in social, interactions. The use of a robotic architecture may strongly reduce the time and effort required to construct a sociable robot. Such architecture must have structures and mechanisms to allow social interaction. behavior control and learning from environment. Learning processes described oil Science of Behavior Analysis may lead to the development of promising methods and Structures for constructing robots able to behave socially and learn through interactions from the environment by a process of contingency learning. In this paper, we present a robotic architecture inspired from Behavior Analysis. Methods and structures of the proposed architecture, including a hybrid knowledge representation. are presented and discussed. The architecture has been evaluated in the context of a nontrivial real problem: the learning of the shared attention, employing an interactive robotic head. The learning capabilities of this architecture have been analyzed by observing the robot interacting with the human and the environment. The obtained results show that the robotic architecture is able to produce appropriate behavior and to learn from social interaction. (C) 2009 Elsevier Inc. All rights reserved.

Educational robotics as a learning aid for disabled children

Severe disabled children have little chance of environmental and social exploration and discovery, and due this lack of interaction and independency, it may lead to an idea that they are unable to do anything by themselves. This idea is called learned helplessness and is very negative for the child cognitive development and social development as well. With this entire situation it is very likely that the self-steam and mood of this child. Trying to help these children on this situation, educational robotics can offer and aid, once it can give them a certain degree of independency in exploration of environment. The system developed in this work allows the child to transmit the commands to a robot. Sensors placed on the child's body can obtain information from head movement or muscle pulses to command the robot to carry the tasks. Also, this system can be used with a variety of robots, being necessary just a previous configuration. It is expected that, with the usage of this system, the disabled children have a better cognitive development and social interaction, balancing in a certain way, the negative effects of their disabilities. © 2011 IEEE.

Robotics learning based on collaborative projects

[EN]Nowadays companies demand graduates able to work in multidisciplinary and collaborative projects. Hence, new educational methods are needed in order to support a more advanced society, and progress towards a higher quality of life and sustainability. The University of the Basque Country belongs to the European Higher Education Area, which was created as a result of the Bologna process to ensure the connection and quality of European national educational systems. In this framework, this paper proposes an innovative teaching methodology developed for the "Robotics" subject course that belongs to the syllabus of the B.Sc. degree in Industrial Electronics and Automation Engineering. We present an innovative methodology for Robotics learning based on collaborative projects, aimed at responding to the demands of a multidisciplinary and multilingual society.

Standard commercial education robotic platform: intermediate solution to teaching robotics

Teaching robotics to students at the beginning of their studies has become a huge challenge. Simulation environments can be an effective solution to that challenge where students can interact with simulated robots and have the first contact with robotic constraints. From our previous experience with simulation environments it was possible to observe that students with lower background knowledge in robotics where able to deal with a limited number of constraints, implement a simulated robotic platform and study several sensors. The question is: after this first phase what should be the best approach? Should the student start developing their own hardware? Hardware development is a very important part of an engineer's education but it can also be a difficult phase that could lead to discouragement and loss of motivation in some students. Considering the previous constraints and first year engineering students’ high abandonment rate it is important to develop teaching strategies to deal with this problem in a feasible way. The solution that we propose is the integration of a low-cost standard robotic platform WowWee Rovio as an intermediate solution between the simulation phase and the stage where the students can develop their own robots. This approach will allow the students to keep working in robotic areas such as: cooperative behaviour, perception, navigation and data fusion. The propose approach proved to be a motivation step not only for the students but also for the teachers. Students and teachers were able to reach an agreement between the level of demand imposed by the teachers and satisfaction/motivation of the students.

Learning joint representations for order and timing of perceptual-motor sequences: a dynamic neural field approach

Many of our everyday tasks require the control of the serial order and the timing of component actions. Using the dynamic neural field (DNF) framework, we address the learning of representations that support the performance of precisely time action sequences. In continuation of previous modeling work and robotics implementations, we ask specifically the question how feedback about executed actions might be used by the learning system to fine tune a joint memory representation of the ordinal and the temporal structure which has been initially acquired by observation. The perceptual memory is represented by a self-stabilized, multi-bump activity pattern of neurons encoding instances of a sensory event (e.g., color, position or pitch) which guides sequence learning. The strength of the population representation of each event is a function of elapsed time since sequence onset. We propose and test in simulations a simple learning rule that detects a mismatch between the expected and realized timing of events and adapts the activation strengths in order to compensate for the movement time needed to achieve the desired effect. The simulation results show that the effector-specific memory representation can be robustly recalled. We discuss the impact of the fast, activation-based learning that the DNF framework provides for robotics applications.

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.

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