This paper describes a three-dimensional terrain mapping and classification technique to allow the operation of mobile robots in outdoor environments using laser range finders. We propose the use of a multi-layer perceptron neural network to classify the terrain into navigable, partially navigable, and non-navigable. The maps generated by our approach can be used for path planning, navigation, ...

The objective of this work is to develop a digitized mammograms’ feature extraction approach using Kohonen’s Self-Organizing Maps (SOM). Once developed, the SOM network will be used as the first processing stage in a breast cancer computer aided diagnosis (CAD) system. Its role will be to offer segmented data as input to a second stage dedicated to the diagnosis task, which will be implemented ...

The problem of classifying objects from their ultrasonic signature for robotic applications is studied in this paper. The system developed utilises the spatial diversity of a four element linear array transducer to enhance classification performance. A signal pre-processing technique employing time domain envelope detection in combination with a multi-layer perceptron neural network has yielded...

In this paper a new PID controller is presented. Its peculiarity consists in the non integer order of the derivative and integral, being this order variable in the range [0,1], and providing two further parameters to the controller. The key element is a Multi-layer Perceptron that evaluate a non integer order integration. The neural network approach allows to overcome some problems related to p...

This paper deals with a novel feature extraction method for text dependent speaker recognition. Four female speakers were used to create a text –dependent database for Malayalam (one of the south Indian languages). Discrete Wavelet Transform was used for feature extraction and artificial neural network was used for machine intelligence. In this work we used emotional features for speaker recogn...

The emission rate of minority atmospheric gases is inferred by a new approach based on neural networks. The new network applied is the multi-layer perceptron with backpropagation algorithm for learning. The identification of these surface fluxes is an inverse problem. A comparison between the new neural-inversion and regularized inverse solutions is performed. The results obtained from the neur...

We present a framework that greatly simplifies the evaluations and analyses for optimization in the complex plane through the use of a generalized definition of analyticity. We show how the gradient, relative (natural) gradient, Newton, and Newton variation updates can be easily derived within the framework. We demonstrate application of the framework for training of a multi-layer perceptron ne...

This study presents the development of a hybrid system consisting of an ensemble of Extended Kalman Filter (EKF) based Multi Layer Perceptron Network (MLPN) and a one-pass learning Fuzzy Inference System using Look-up Table Scheme for the recognition of electrocardiogram (ECG) signals. This system can distinguish various types of abnormal ECG signals such as Ventricular Premature Cycle (VPC), T...

This paper introduces NFQ, an algorithm for efficient and effective training of a Q-value function represented by a multi-layer perceptron. Based on the principle of storing and reusing transition experiences, a model-free, neural network based Reinforcement Learning algorithm is proposed. The method is evaluated on three benchmark problems. It is shown empirically, that reasonably few interact...

This paper deals with the compensation for the nonlinear distortion introduced by power-e cient ampli ers on linear modulations by means of equalization. We propose a new equalizer based on a reduced-complexity network called GCMAC. The GCMAC-based equalizer is compared with other well-known structures such as the Volterra lter and the Multi-layer Perceptron. Extensive computer simulations have...