Colin Flanagan

Neural network control of underwater vehicles

In this paper, recent research efforts in the field of the application of neural networks (NNs) for the control of (semi-)autonomous underwater vehicles are reviewed. Based on a literature review the authors propose a classification of approaches to control underwater vehicles using NNs and the presented articles are categorized according to the identified categories. Based on practical results as described in the discussed literature this paper presents a qualitative assessment regarding the performance of the control strategies. Per category, or control strategy, the major advantages and disadvantages are identified and discussed.

Interpreting complex data from a three-sensor multipoint optical fibre ethanol concentration sensor system using artificial neural network pattern recognition

A three-sensor element multipoint optical fibre sensor system capable of detecting varying ethanol concentrations in water for use in industrial process water systems is reported. The sensor system utilizes a U-bend configuration for each sensor element in order to maximize the sensitivity of each of the sensing regions along the optical fibre cable. The sensor system is interrogated using a technique known as optical time domain reflectometry, as this method is capable of detecting attenuation over distance. Analysis of the data arising from the sensor system is performed using artificial neural network pattern recognition techniques, coupled with Fourier-transform-based signal processing. The signal processing techniques are applied to the obtained sensor system data, prior to the artificial neural network analysis, with the aim of reducing the computational resources required by the implemented artificial neural network.

Real-time, non-intrusive evaluation of VoIP

Speech quality, as perceived by the users of Voice over Internet Protocol (VoIP) telephony, is critically important to the uptake of this service. VoIP quality can be degraded by network layer problems (delay, jitter, packet loss). This paper presents a method for real-time, non-intrusive speech quality estimation for VoIP that emulates the subjective listening quality measures based on Mean Opinion Scores (MOS). MOS provide the numerical indication of perceived quality of speech. We employ a Genetic Programming based symbolic regression approach to derive a speech quality estimation model. Our results compare favorably with the International Telecommunications Union-Telecommunication Standardization (ITU-T) PESQ algorithm which is the most widely accepted standard for speech quality estimation. Moreover, our model is suitable for real-time speech quality estimation of VoIP while PESQ is not. The performance of the proposed model was also compared to the new ITU-T recommendation P.563 for non-intrusive speech quality estimation and an improved performance was observed.

Design of a system that uses optical-fiber sensors and neural networks to control a large-scale industrial oven by monitoring the food quality online

An optical-fiber sensor-based system has been designed to assist in the controlling of a large-scale industrial by monitoring the color of the food product being cooked. The system monitors the color of the food as it cooks by examining the reflected visible light, from the surface and/or core of the cooked product. A trained backpropagation neural network acts as a classifier and is used to interpret the extent to which each product is cooked with regard to the aesthetics of the food. Principal component analysis is also included before the neural network as a method of feature extraction. This is implemented using Karhunen-Loeve decomposition. A wide range of food products have been examined and accurately classified, demonstrating the versatility and repeatability of the system over time. These products include minced beef burgers and steamed chicken fillets.

An optical fibre ethanol concentration sensor utilizing Fourier transform signal processing analysis and artificial neural network pattern recognition

An optical fibre sensor, which is capable of detecting varying percentages of ethanol in water, is reported. In order to maximize sensitivity, a U-bend configuration was used for the sensor where the cladding was removed and the core exposed directly to the fluid under test. The sensor was interrogated using optical time domain reflectometry (OTDR). OTDR is chosen as it is a recognized technique for the interrogation of distributed multipoint sensors and it is intended to extend this work to multiple sensors on a single fibre in the future. In this investigation the sensor was exposed to 12.5, 25 and 50% ethanol and distilled water. The signal processing technique has been designed to optimize the neural network adopted in the existing sensor system. In this investigation a discrete Fourier transform, using a fast Fourier transform algorithm, is chosen and its application leads to an improvement in efficiency of the neural network, i.e. reducing the number of input and hidden layer nodes required by the artificial neural network. Using a Stuttgart neural network simulator, a feed-forward three-layer neural network was constructed with the aim of successfully classifying the sensor test conditions based on the frequency domain response of the sensor.

An optical-fiber sensor for use in water systems utilizing digital signal processing techniques and artificial neural network pattern recognition

An optical-fiber sensor is reported which is capable of detecting ethanol in water. A single optical-fiber sensor was incorporated into a 1-km length of 62.5-/spl mu/m core diameter polymer-clad silica optical fiber. In order to maximize sensitivity, a U-bend configuration was used for the sensor where the cladding was removed and the core exposed directly to the fluid under test. The sensor was interrogated using optical time domain reflectrometry, as it is intended to extend this work to multiple sensors on a single fiber. In this investigation, the sensor was exposed to air, water, and alcohol. The signal processing technique has been designed to optimize the neural network adopted in the existing sensor system. In this investigation, a discrete Fourier transform, using a fast Fourier transform algorithm, is chosen and its application leads to an improvement in efficiency of the neural network i.e., minimizing the computing resources. Using the Stuttgart neural network simulator, a feed-forward three-layer neural network was constructed with the number of input nodes corresponding to the number of points required to represent the sensor frequency domain response.

A multi-point optical fibre sensor for condition monitoring in process water systems based on pattern recognition

Preliminary results are presented for a multi-point optical fibre sensor designed to detect the presence of chemical species in water at spatial intervals of greater than 20 m. The sensor is addressed using optical time domain reflectometry (OTDR) with a spatial resolution of 10 m. The optical signals arising from the OTDR are highly complex due to interfering effects from external parameters such as localised fibre straining and temperature changes. Because of this level of complexity it has been found advantageous to use artificial neural networks (ANNs) as classifiers on the OTDR signals. The preliminary system has been trained initially to recognise only the presence of water, although it is planned to extend this capability to recognise the presence of contaminants in the water such as bacteria and chemical pollutants. Initial investigations show that different contaminants and interfering parameters (cross-sensitivities) may give rise to characteristic signatures on the OTDR signal which may be identified by the pattern recognition software.

Neural network augmented identification of underwater vehicle models

In this article the use of neural networks in the identification of models for underwater vehicles is discussed. Rather than using a neural network in parallel with the known model to account for unmodelled phenomena in a model wide fashion, knowledge regarding the various parts of the model is used to apply neural networks for those parts of the model that are most uncertain. As an example, the damping of an underwater vehicle is identified using neural networks. The performance of the neural network based model is demonstrated in simulations using the neural networks in a feed forward controller. The advantages of online learning are shown in case of noise impaired measurements and changing dynamics due to a change in toolskid.

A neural networks based approach for determining fouling of multi-point optical fibre sensors in water systems

An optical fibre sensor system for monitoring contamination in water supplies is presented. The sensor comprises a number of individual sensor elements on a single fibre loop. It is addressed using optical time domain reflectometry so that the required spatial resolution of 1 metre is achieved. Analysis of the signals at the receiving end is performed using artificial neural networks coupled with pattern recognition techniques, thus allowing external influences such as the degree of sensor fouling to be detected. In this investigation limescale build-up in hard water is investigated as the interfering parameter.

Evaluation of Asynchronous Event Mechanisms for Browser-based Real-time Communication Integration

Desktop based real-time communication applications are commonly used for presence based instant messaging and telephony applications. Such applications use installed desktop components to handle real-time asynchronous events on the client originating from the communication system. When the client is based in a web browser these installed components are not available however browser-based communication applications are required to handle the same type of asynchronous events. Moreover, browser-based clients which typically run over HTTP are challenged by HTTP itself which is designed to be a synchronous request-response protocol. We contend that a suitable mechanism can be found to deliver asynchronous real-time events to browser-based applications