Indoor Localization System based on Artificial Landmarks and Monocular Vision

Andry Maykol G. Pinto, A. Paulo Moreira, Paulo G. Costa
2012 TELKOMNIKA (Telecommunication Computing Electronics and Control)  
Abstrak Makalah ini menyajikan pendekatan lokalisasi visual yang cocok untuk lingkungan rumah tangga dan industri karena akurat, handal dan estimasi pandangan yang baik. Robot mobile dilengkapi dengan kamera tunggal yang meng-update pandangan kapanpun ketika terdapat landmark. Inovasiyang disajikan dalam penelitian ini difokuskan pada sistem landmark buatan yang memiliki kemampuan untuk mendeteksi keberadaan robot karena kedua entitas berkomunikasi satu sama lain menggunakan protokol sinyal
more » ... a merah termodulasi frekuensi. Selain ini kemampuan komunikasi, setiap landmark memiliki beberapa LED intensitas tinggi yang bersinar hanya untuk beberapa kasus sesuai dengan komunikasi, yang memungkinkan bekerjanya alat pengatur cahaya kamera dan berkedipnya LED untuk sinkronisasi. Sinkronisasi ini meningkatkan toleransi sistem tentang perubahan kecerahan pada lampu ambien dari waktu ke waktu, secara independen dari lokasi landmark. Oleh karena itu, langit-langit lingkungan yang dihuni dengan beberapa landmark dan Extended Kalman Filter digunakan untuk menggabungkan perhitungan mati dan informasi landmark. Hal ini meningkatkan fleksibilitas sistem dengan mengurangi jumlah landmark yang diperlukan. Evaluasi eksperimental dilakukan dalam lingkungan indoor nyata dengan purwarupa kursi roda otonom. Kata kunci: beacon aktif, landmark buatan, lokalisasi visual, navigasi robot Abstract This paper presents a visual localization approach that is suitable for domestic and industrial environments as it enables accurate, reliable and robust pose estimation. The mobile robot is equipped with a single camera which update sits pose whenever a landmark is available on the field of view. The innovation presented by this research focuses on the artificial landmark system which has the ability to detect the presence of the robot, since both entities communicate with each other using an infrared signal protocol modulated in frequency. Besides this communication capability, each landmark has several high intensity light-emitting diodes (LEDs) that shine only for some instances according to the communication, which makes it possible for the camera shutter and the blinking of the LEDs to synchronize. This synchronization increases the system tolerance concerning changes in brightness in the ambient lights over time, independently of the landmarks location. Therefore, the environment's ceiling is populated with several landmarks and an Extended Kalman Filter is used to combine the dead-reckoning and landmark information. This increases the flexibility of the system by reducing the number of landmarks required. The experimental evaluation was conducted in a real indoor environment with an autonomous wheelchair prototype. 610 This article addresses the problem of a generic, robust and accurate indoor localization system based on artificial landmarks and a single camera. Similarly to the explorers who used the stars to navigate their boats between continents, the approach presented here is based on a constellation of landmarks placed on the ceiling. This will allow several robots to simultaneously locate themselves. The presented method includes two phases: mapping and localization. This article focuses on the localization process since it represents what is new and what is different from the approaches presently described in the literature. The mapping creates the prior knowledge on the position of each landmark in the environment, for example, a common simultaneous localization and mapping (SLAM) technique can be used in this step in order to automatically generate relations between landmark positions. After the mapping stage, the robot can move freely in the environment, using the current visual information to determine position and orientation. The core of this localization method is the "intelligent" landmark, which is called Synchronized Color Landmark (SCL) in this paper as this is composed of a cluster of RGB LEDs (with different colors) which shine synchronized with the camera shutter, using a communication protocol that resorts to a frequency modulated infrared signal whenever the robot is close. The high intensity RGB LEDs and the synchronization made the system highly immune to external light sources, which represents an enormous advantage comparatively to other similar methods. This concept, called SincroVision (patented),is under development at the Faculty of Engineering of the University of Porto. The aim is to automatically teach robotic manipulators [1] . The system is evaluated considering its robustness and accuracy in a robot platformautonomous wheelchair, IntelWheels [4] [5] [6] , and in realistic testing scenarios. The results were obtained with a real person sitting on the wheelchair, which represents a drawback for the localization accuracy due to the unbalanced weight distribution and movements made by the person. Given the intermittent availability of the landmarks, an Extended Kalman Filter (EKF)is used to update the dead-reckoning estimation. One major contribution of this research is the remarkable accuracy and robustness of the localization system, even in a presence of illumination instabilities. This paper is organized as follows: the next section introduces a brief presentation about some related works. Section 3 provides a detailed explanation about the localization concept. Section 4 analyzes the robustness and accuracy performances, and finally, Section 5presents the main conclusions and topics for future research.
doi:10.12928/telkomnika.v10i4.848 fatcat:btsfw6oiird75cffuxleuxrr3a