Sistema de posicionamiento local para localización absoluta de robots móviles en espacios inteligentes mediante infrarrojos

  1. MARTIN GOROSTIZA, ERNESTO
Supervised by:
  1. José Luis Lázaro Galilea Director
  2. Francisco Javier Meca Meca Co-director

Defence university: Universidad de Alcalá

Fecha de defensa: 30 January 2012

Committee:
  1. Jesús Ureña Ureña Chair
  2. Daniel Pizarro Pérez Secretary
  3. Paloma Rodríguez Horche Committee member
  4. Eduardo Martos Naya Committee member
  5. José Luis Martínez Lastra Committee member
Department:
  1. Electrónica

Type: Thesis

Teseo: 320491 DIALNET

Abstract

This Thesis aims at developing an Infrarred LPS (Local Positioning System) for indoor mobile-robotics applications, in the framework of the Intelligent Spaces. The Intelligent Spaces are expected to grow in the near future, meaning a research challenge, as a lot of new applications are envisaged in this scope. In the LPS proposed in this thesis, the position of an Infrarred (IR) emitter, mounted onboard a mobile robot, is obtained by means of hyperbolic trilateration.. While most of these types of systems are based on Ultrasound (US) and Artificial Vision or RF, IR has been less explored as an alternative for LPSs. The challenge with IR is achieving an acceptable accuracy (in the range of 1-10cm), as 1µs in time-error would mean 30cm in distance-error (US provides around 1cm-accuracy and artificial vision achieves accuracy in a 4-10cm interval). To reach accuracy, high signal to noise ratio (SNR) is needed but, at the same time, wide emission-angle is needed too, in order to have acceptable coverage area and reach a minimum of three detectors (with fixed and known emitter height) at once, for trilateration. However, for a fixed total emitted power, the more the angle is opened, the less the power per solid-angle unit is received by the detector. On the other hand, long integrating-time enhances precision but worsens the system response-time, which must meet robot speed requirements. Thus, in the conception of the system, there is a severe trade-off between SNR, angle, and response-time, or, equivalently, between precision, coverage and mobile robot speed. In this context, and IR-LPS for mobile robots has been proposed, based on hyperbolic trilateration with differential phase-sift measurements, which provides accuracy better than 5cm. The system is an alternative approach to the existing solutions based on other technologies different from IR, and also to most of the IR localization systems, which work with laser. The measurement and localization method, based on differential-phase-shift measurements, also represents a different approach compared to the methods used by the majority of the systems, based either on time-of-flight measurements, or on received signal strength intensity