Search

[EN] The management of people and vehicles mobility is an aspect of continuous study due to its contribution to pollution. Traffic light control determines the queues that can form at crossroads. Usually, this control is not adapted to the existing traffi at a specific time since the adaptation implies knowing the pedestrians and vehicles circulating at all times. This article proposes using modular intelligent devices that allow vehicles to be detected and access times to the intersection to be changed depending on the circumstances. A simulation has been carried out generating loads in MatLab and simulating the control with Simulink. A traffic light cycle with fixed times has been simulated and compared with cycles with varying service times depending on a load of pedestrians and vehicles. In the article, the Op and Sat indicators are proposed to measure the optimisation of the control algorithm on the crossing state. Using these indicators, it has been shown that it is possible to optimise the waiting time by 50%, almost independently of the traffic load in the best case., [ES] La gestión de la movilidad de personas y vehículos es un aspecto de continuo estudio debido a la relevancia que tiene en la contribución a la polución. El control de los semáforos determina las colas que en los cruces se pueden formar. Habitualmente este control no está adaptado al tráfico existente en un momento concreto, dado que la adaptación implica conocer los peatones y vehículos que se encuentran circulando en cada momento. Para resolver este problema, en el artículo se propone el uso de unos dispositivos inteligentes modulares que permiten detectar los vehículos y cambiar los tiempos de acceso al cruce dependiendo de las circunstancias. Para validar el sistema se ha realizado una simulación generando cargas en MatLab y simulando el control con Simulink. Se ha simulado un ciclo de semáforo con tiempos fijos y se ha comparado con ciclos de tiempos variables en función de la carga de peatones y de vehículos. En el artículo se proponen los indicadores Op y Sat como método de medición de la optimización del algoritmo de control sobre el estado del cruce. Por medio de dichos indicadores se ha comprobado que en el mejor de los casos es posible optimizar en un 50 % el tiempo de espera de forma casi independiente de la carga de tráfico., Ministerio de Ciencia e Innovación de España Proyecto MICINN: CICYT PRECON-I4: Sistemas informáticos predecibles y confiables para la Industria 4.0 TIN2017-86520-C3-1-R

[EN] Robot behavior definition aims to classify and specify the robot tasks execution. Behavior architecture design is crucial for proper robot operation performance. According to this, this work aims to establish a robot behavior architecture based on distributed intelligent services. Therefore, behavior definition is set in a high-level delegating the task execution to distributed services provided by network abstractions characterized as Smart Resources. In order to provide a mechanism to measure the performance of this architecture, an evaluation mechanisms based on a service performance composition is introduced. In order to test this proposal it is designed a real use case implementing the proposed robot behavior architecture on a real navigation task., Work supported by the Spanish Science and Innovation Ministry MICINN: CICYT 866 project M2C2: Codiseño de sistemas de control con criticidad mixta basado en 867 misiones TIN2014-56158-C4-4-P and PAID (Polytechnic University of Valencia): 868 UPV-PAID-FPI-2013.

[EN] Mobile robots execute complexes tasks that involve the management of several embedded sensors and actuators. Therefore, in many cases, a robot is characterized as an intelligent distributed system formed with a central unit, which manages the on-board embedded devices and distributes the tasks execution. Embedded devices are also evolving to more complex systems. These systems are developed not only for executing simple tasks but also for offering some advanced mechanisms. Thus, complex data processing, adaptive execution, or fault-tolerance routines are some common system features. The Smart Resource topology has been developed in order to manage these embedded systems. This topology offers high-level routines that rely on a certain physical hardware execution. Therefore, Smart Resources are defined as distributed services providers, which operates within some context and quality requirements. Provided services can adapt its execution in order accomplish the set requirements and maximize the system performance. How to improve the versatility of the Smart Resources by making their services compatibles with the Robot Operating System (ROS) is addressed along this work. This solution integrates all the execution mechanisms provided by ROS with the service distribution, adaptive execution, and fault-tolerance routines offered by the Smart Resources. This integration is tested through a set of experiments using the Turtlebot robot platform and a simulated version of it. In both approaches ROS mechanisms are used to access the Smart Resource Services. Finally, obtained results are used to characterize the performance of this proposal., Work supported by the Spanish Science and Innovation Ministry MICINN: CICYT project M2C2: "Codiseno de sistemas de control con criticidad mixta basado en misiones" TIN2014-56158-C4-4-P and PAID (Polytechnic University of Valencia): UPV-PAID-FPI-2013.

This article presents the experimental work developed to test the viability and to measure the efficiency of the intelligent control distributed architectures. To do this, a simulated navigation scenario of Braitenberg vehicles has been developed. To control the vehicles, system uses a distributed control architecture that provides support to QoS and QoC parameters to optimize de system. The architecture uses a Publish-Subscribe model, based on Data Distribution Service to send the control messages. Due to the nature of the Publish-Subscribe model, the architecture is suitable to implement event-based control systems. The architecture has been called FSACtrl. To test the efficiency, the architecture provides a set of Quality of Service parameters. In the experiment described in this paper, the performance is used as a reference. The measuring of the quality of the navigation is done through the Integrated Time Average Error as Quality of Control parameter. Tested scenarios are: an environment without quality parameter managing, an environment with a relevant message filtering and an environment with a predictive filtering determined by the type of scenario used. Results obtained show that some of the processing performed in the control nodes can be moved to the middleware to optimize the robot navigation.