Advanced techniques for the prediction originating from conventional neural network approaches, the local risk minimizer introduced by Vapkin, showed that it consistently outperforms other conventional neural network ...
详细信息
Advanced techniques for the prediction originating from conventional neural network approaches, the local risk minimizer introduced by Vapkin, showed that it consistently outperforms other conventional neural network techniques. An ESPRIT project is presented to take advantage of the techniques described to perform efficient prediction in demanding finance applications. The conventional multi-layer perception and Vapkin's theory of local estimation are compared for the applications of prediction: one in a check processing center and the other, in a Credit Card Call Center, both operating at sites run by SLIGOS.
The need for providing gas turbine engine control with more sophisticated controls that reduce cost is discussed. Interest is given in the exploitation of Commercial-Off-The-Shelf (COTS) technologies reduces size, cos...
详细信息
The need for providing gas turbine engine control with more sophisticated controls that reduce cost is discussed. Interest is given in the exploitation of Commercial-Off-The-Shelf (COTS) technologies reduces size, cost and weight will provide for this need of controlling gas turbine engines. A number of key COTS technologies are identified: increased miniaturization using multichip modules (MCM);use of COTS databases from automotive applications;use of non-MIL SPEC components and high temperature electronics;and smart actuators and sensors. A review of the technologies and approaches that highlights the need to change the Full Authority Digital Engine controller (FADEC) architecture is presented.
An alternative approach using neural networks is presented to generate realistic bounds that considers two aspects of data which are known to influence the resultant prediction error. This is the ability of the networ...
详细信息
An alternative approach using neural networks is presented to generate realistic bounds that considers two aspects of data which are known to influence the resultant prediction error. This is the ability of the network to predict the output and impact of the density of the training data on the resultant bounds. Neural network predictions with confidence bounds provide a more robust and believable representation of the forecast, and provide industries with greater confidence when using industrial process applications. The power of the technique lies in the fact that it is not necessary to identify an underlying distribution which is to satisfy and is equally applicable to all forms of network topologies and non-linear modeling procedures.
We summarize the results of two applications of soft-computing to walking robots. In the first case a fuzzy controller has been adopted to solve a classical problem of controlling the position of three DC motors. Such...
详细信息
We summarize the results of two applications of soft-computing to walking robots. In the first case a fuzzy controller has been adopted to solve a classical problem of controlling the position of three DC motors. Such a controller showed good dynamical performance, was very easy to be tuned and allowed one to add ad-hoc rules in order to change its behaviour. In the second case soft computing techniques allowed one to solve the problem of controlling the trajectory of a new kind of uncertain micro-walking robots.
This report presents a model that covers a wide range of mobileapplications. The model is based on the electronic equivalent of the Post-it note, which is called stick-e note. It is hoped that the stick-e model achie...
详细信息
This report presents a model that covers a wide range of mobileapplications. The model is based on the electronic equivalent of the Post-it note, which is called stick-e note. It is hoped that the stick-e model achieves generality and simplicity.
Engineers and scientists working in mobilecomputing often have to justify or even defend their efforts when dealing with people from industry. The developer's enthusiasm is regularly challenged by responses of th...
详细信息
Engineers and scientists working in mobilecomputing often have to justify or even defend their efforts when dealing with people from industry. The developer's enthusiasm is regularly challenged by responses of the form 'Yes, but who is actually going to use it'. This talk attempts to move from technically justifying work in mobilecomputing to commercially motivating it. Collaborative mobileapplications have a role to play in the utility environment and this role is set to expand as competition develops.
mobile end-users will exist in environments subject to rapid and marked change. In order to make best use of the available resources, applications and system services will have to adapt to these changes. In this artic...
详细信息
mobile end-users will exist in environments subject to rapid and marked change. In order to make best use of the available resources, applications and system services will have to adapt to these changes. In this article, a prototype platform developed at Lancaster is described. This platform is designed to support this process of adaptation.
This paper is concerned with a number of techniques for improving the performance of the ETSI HIPERLAN physical layer. After a brief review of the HIPERLAN standard, a description of pre-coded GMSK is presented. The m...
详细信息
This paper is concerned with a number of techniques for improving the performance of the ETSI HIPERLAN physical layer. After a brief review of the HIPERLAN standard, a description of pre-coded GMSK is presented. The methods used to model the HIPERLAN radio channel are introduced and the performance impact of combined diversity and Decision Feedback Equalisation (DFE) presented. Important issues for an adhoc radio LAN such as frame synchronisation and frequency offset correction are addressed from a practical standpoint. Finally, incorporating the standard's BCH(31,26) block code and interleaving strategy, the Bit Error Rate (BER) and Frame Error Rate (FER) performance is simulated both for standard and enhanced receiver designs.
The proceedings contains 8 papers from the ieecolloquium on intelligent measuring systems for controlapplications. Some of the topics discussed include: inferential estimation;industrial process control;Analytical P...
详细信息
The proceedings contains 8 papers from the ieecolloquium on intelligent measuring systems for controlapplications. Some of the topics discussed include: inferential estimation;industrial process control;Analytical Process control;close-coupled robot;sewer flow mean velocity measurements;ultrasonic velocity transducers;cluster analysis techniques;multisensor data fusion;decentralized architecture;cooperative arm tasks;fault diagnosis and fault detection.
The proceedings contains 10 papers. Topics discussed include case based reasoning, prototypes in case based diagnosis, application of case based reasoning in object oriented database, information retrieval, applicatio...
详细信息
The proceedings contains 10 papers. Topics discussed include case based reasoning, prototypes in case based diagnosis, application of case based reasoning in object oriented database, information retrieval, application in accounting, case based cost estimation and design.
暂无评论