Tracking and data association
A hierarchical approach to line extraction based on the Hough transform
Computer Vision, Graphics, and Image Processing
Motion planning with uncertainty: a landmark approach
Artificial Intelligence - Special volume on planning and scheduling
Neural networks for mobile robot piloting control
Neural network for robotic control
Directed Sonar Sensing for Mobile Robot Navigation
Directed Sonar Sensing for Mobile Robot Navigation
A Genetic Algorithm for Mobile Robot Localization Using Ultrasonic Sensors
Journal of Intelligent and Robotic Systems
Hi-index | 0.00 |
This article describes a system for locating a mobile robot moving in a known indoor environment. The simplest method for position estimation of a mobile robot is odometry; but, as the robot moves errors of estimation are accumulated and the accuracy of the estimation decreases. The problem is solved by relocalizing periodically the vehicle by means of external references. Besides the common technique of obtaining the position estimation from odometry with observations of landmarks, a fuzzy perception planner, that actively supervises the data acquisition and landmark extraction, is proposed. An Extended Kalman Filter is normally used to correct the position and orientation of the vehicle from the difference between the observed distances and angles to each landmark and the estimated ones. This article presents a localization system which uses visual detection of natural and artificial landmarks for relocalization. Artificial landmarks are solid circles placed on the walls and natural landmarks are nameplates fixed at the entrance of each room. An important goal of the system is to detect the landmarks with the vehicle in motion. That is not easy because of the uncertainty of the robot's position, the unpredictability of the vehicle's movements and the time necessary to move the camera. All these parameters are taken into account in the proposed system.