IngentaConnect Mobile robot localization using a non-linear evolutionary filter

Authors: Armingol, J.M.; de la Escalera, A.; Moreno, L.; Salichs, M.A.

Source: Advanced Robotics, Volume 16, Number 7, 2002 , pp. 629-652(24)

Abstract:

This article describes a localization system for autonomous mobile robot navigation in an indoor semi-structured environment. A peripheral ring of 24 ultrasonic sensors and a camera with a motorized zoom on a pan-tilt platform are used to obtain the information required for the localization process. A non-linear filter based on a genetic algorithm as an emerging optimization method to search for optimal positions is presented. The proposed algorithm is based upon an iterative extended Kalman filter (EKF), which utilizes matches between observed geometric beacons and a generic map of beacon locations and the detection of artificial landmarks, to correct the position and orientation of the vehicle. No exhaustive map of the environment is provided to the mobile robot. It must work with a generic description of the kinds of entities in the environment. The resulting self-localization module has been integrated successfully in a more complex navigation system based on a reactive architecture. Various experimental results show the effectiveness of the presented algorithm, including a comparison with the EKF method.

Keywords: MOBILE ROBOTS; GENETIC ALGORITHMS; KALMAN FILTER; ULTRASONIC SENSORS; VISUAL LOCALIZATION

Document Type: Research article

DOI: 10.1163/15685530260390755

The full text electronic article is available for purchase. You will be able to download the full text electronic article after payment.

$25.00 plus tax