Відкритий доступ від SAGE Journals з нагоди проведення Європейського форуму з робототехніки: Reliable confirmation of an object identity by a mobile robot


SAGE Journals з нагоди проведення Європейського форуму з робототехніки (European Robotics Forum 2016) продовжує відкривати доступ до окремих наукових статей.


Reliable confirmation of an object identity by a mobile robot: A mixed appearance/localization-driven motion approach

  1. Israel Becerra (Centro de Investigación en Matemáticas, CIMAT, México)
  2. Luis M Valentín-Coronado (Centro de Investigación en Matemáticas, CIMAT, México)
  3. Rafael Murrieta-Cid (Centro de Investigación en Matemáticas, CIMAT, México)
  4. Jean-Claude Latombe (Artificial Intelligence Laboratory, Computer Science Department, Stanford University, USA)

Abstract

This paper investigates the problem of confirming the identity of a candidate object (expected to be a target based on some crude visual clues) with a mobile robot equipped with visual sensing capabilities. We present a method whose main novelty is to mix localization of the robot relative to the candidate object and to confirm that it is the sought target. This twofold approach drastically reduces false positives. Identity confirmation with this twofold goal is modeled as a Partially-Observable Markov Decision Process, where the states are the cells of the space decomposition. It is solved using Stochastic Dynamic Programming with imperfect state information. A robotic system using this method has been implemented and tests have been carried out both in simulation and with a real robot. The experiments empirically validate the use of various metrics, and demonstrate their ability to perform well in different settings.
  • Object identification
  • motion strategies
  • appearance-guided motion
  • robot localization
Повний текст: 
Reliable confirmation of an object identity by a mobile robot: A mixed appearance/localization-driven motion approach



Коментарі

Приєднуйтесь до нас в соціальних мережах

Icon Icon

Книжкова полиця видань НБ ЗНТУ

Популярні публікації

Загальна кількість переглядів сторінки