Sciweavers

Share
IWANN
2001
Springer

Landmark Recognition for Autonomous Navigation Using Odometric Information and a Network of Perceptrons

9 years 4 months ago
Landmark Recognition for Autonomous Navigation Using Odometric Information and a Network of Perceptrons
In this paper two methods for the detection and recognition of landmarks to be used in topological modeling for autonomous mobile robots are presented. The first method is based on odometric information and the distance between the estimated position of the robot and the already existing landmarks. Due to significant errors arising in the robot’s position measurements, the distance-based recognition method performs quite poorly. For such reason a much more robust method, which is based on a neural network formed by perceptrons as the basic neural unit is proposed. Apart from performing very satisfactorily in the detection and recognition of landmarks, the simplicity of the selected ANN architecture makes its implementation very attractive from the computational standpoint and guarantees its application to real-time autonomous navigation.
Javier de Lope Asiaín, Darío Maraval
Added 30 Jul 2010
Updated 30 Jul 2010
Type Conference
Year 2001
Where IWANN
Authors Javier de Lope Asiaín, Darío Maravall Gómez-Allende
Comments (0)
books