Métodos y herramientas para robótica ubicua integrando robots móviles y sensores distribuidos

• Autores: Adrián Jiménez González
• Directores de la Tesis: José Ramiro Martínez de Dios (dir. tes.), Aníbal Ollero Baturone (codir. tes.)
• Lectura: En la Universidad de Sevilla ( España ) en 2013
• Idioma: español
• Tribunal Calificador de la Tesis: Alberto Sanfeliu Cortés (presid.), Rubén Martín Clemente (secret.), Jesús Manuel Gómez de Gabriel (voc.), Pedro José Marrón Chicote (voc.), José Guillermo Heredia Benot (voc.)
• Enlaces
  • Tesis en acceso abierto en: TESEO
• Resumen

  • Recently, robotic technologies have been integrated with a variety of technological disciplines in the context of ubiquitous systems such as ubiquitous and pervasive computing, sensor networks, and ambient intelligence. This dissertation describes methods and tools to enable cooperation between sensor networks and mobile robots within an ubiquitous robotics context. More concretely, it provides solutions to deal with the high heterogeneity intrinsically present in ubiquitous robotics systems. This Ph.D. Thesis proposes a cooperative calibration framework for heterogeneous sensor measurements within an ubiquitous system; it presents an integrating software architecture for interoperability among heterogeneous ubiquitous devices; and it describes the implementation of an experimental tool for ubiquitous robotics in order to palliate the lack of suitable integrated testbeds for ubiquitous robotics.

    The research work presented in this thesis has been built on the groundings of an extensive survey of ubiquitous robotics systems. The main conclusions of the survey include a proposal of taxonomy according to different criteria such as heterogeneity and interoperability among technologies and the identification of existing gaps, hot research issues and future trends. As a second contribution, this thesis proposes a cooperative calibration framework to improve the robustness and adaptivity of distributed and heterogeneous sensors within an ubiquitous robotics system. This framework has been instantiated for the cooperative calibration of static and mobile camera networks subject to strong changes in lighting conditions. A consensus-based multi-camera cooperative calibration algorithm enables all cameras in the system to share lighting models of the objects in the scenario. This framework has been also instantiated to improve the robustness of RSSI based localization for ubiquitous systems subject to changes in transmission power of anchor WSN nodes. The main objective of the proposed method is to improve the accuracy and adaptability of RSSI fingerprinting in indoors localization.

    Furthermore, a novel architecture for ubiquitous robotics systems is proposed. Heterogeneity abstraction, interoperability, modularity and flexibility are the main requirements for this software architecture. The solution adopted is to use an integrating layer through which all the modules intercommunicate in a peer-to-peer basis using standardized interfaces. This architecture has been applied to the development of a general-purpose ubiquitous robotics testbed comprised of a Wireless Sensor Network, a static camera network and a set of 5 mobile robots which carry WSN node and a camera (different from those in the static camera network). The objective of the implemented testbed is to allow a wide range of experiments integrating mobile robots and Wireless Sensor Networks. The testbed has been used by users from worldwide institutions, which have performed experiments in the main research lines within the ubiquitous robotics discipline.