Comparative analysis of different approaches to target differentiation and localization with sonar

Abstract

This study compares the performances of different methods for the differentiation and localization of commonly encountered features in indoor environments. Differentiation of such features is of interest for intelligent systems in a variety of applications such as system control based on acoustic signal detection and identification, map building, navigation, obstacle avoidance, and target tracking. Different representations of amplitude and time-of-2ight measurement patterns experimentally acquired from a real sonar system are processed. The approaches compared in this study include the target differentiation algorithm, Dempster-Shafer evidential reasoning, different kinds of voting schemes, statistical pattern recognition techniques (k-nearest neighbor classifier, kernel estimator, parameterized density estimator, linear discriminant analysis, and fuzzy c-means clustering algorithm), and artificial neural networks. The neural networks are trained with different input signal representations obtained usingpre-processing techniques such as discrete ordinary and fractional Fourier, Hartley and wavelet transforms, and Kohonen's self-organizing feature map. The use of neural networks trained with the back-propagation algorithm, usually with fractional Fourier transform or wavelet pre-processing results in near perfect differentiation, around 85% correct range estimation and around 95% correct azimuth estimation, which would be satisfactory in a wide range of applications. © 2002 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved

A Study on the De-Noising of the Passive Sonar in Underwater Robot Based on the Gabor

“…We can get ships radiated noise with formula (1) .Considering the propagation loss, we take the spherical spread loss formula [9] In the formula, R is the distance from ship to sonar, a is the sound absorption coefficient. With R=2Km,f=2KHz, Ship passive sonar signal from propagation loss can be shown as Figure 6.…”

Fractional Transforms in Optical Information Processing

“…Adjusting the fractional order of the fractional FT of the input signal leads to an overall improvement of the neural network performance, as has been demonstrated on the example of recognition and position estimation of different objects from their sonar returns. In [92], a comparative analysis has been made of different approaches of target differentiation and localization, including the target differentiation algorithm, Dempster-Shafer evidential reasoning, different kinds of voting schemes, statistical pattern recognition techniques (the knearest neighbor classifier, the kernel estimator, the parameterized density estimator, linear discriminant analysis, and the fuzzy c-means clustering algorithm), as well as artificial neural networks, trained with different input signal representations obtained using preprocessing techniques such as discrete ordinary and fractional Fourier, Hartley and wavelet transforms, and Kohonen's self-organizing feature map. It has been shown that the use of neural networks trained by the back-propagation algorithm with fractional FT preprocessing results in near-perfect differentiation, around 85% correct range estimation, and around 95% correct azimuth estimation.…”

Contribución a la autolocalización de robots móviles basada en la fusión de información multisensorial.

“…No obstante, al utilizar este tipo de sensores hay que tomar en cuenta ciertas 1.2. OBJETIVOS limitaciones como por ejemplo que el sensor de ultrasonidos tiene una resolución angular muy pobre(típicamente 25 grados) [16], o que el sensor de infrarrojos ofrece poca precisión y fiabilidad en la medición de distancias(aproximadamente 10 %) [18]. Estos razonamientos, aunado al hecho de que los sensores de bajo coste en general son poco fiables, obligan a desarrollar metodologías mediante las cuales se puedan fusionar la información proveniente de múltiples sistemas sensoriales, a fin de hacerle frente a los diversos retos de localización y modelado del entorno a los cuales se enfrenta un robot móvil con características de autonomía.…”

“…Basándose en la figura 4.28, se determina que a partir de los datos de posición del robot x, y, θ y de los datos de distancia de los sensores de infrarrojos 16 , la proyección de cada uno de los puntos reflectores p = x p , y p en el plano cartesiano global, se puede calcular de acuerdo a las ecuaciones…”