Low cost vector supercomputer
Based on high-end graphics cards with currently up to 240 SPs (streaming processors) vector supercomputers with a maximum of 960 SPs will be built for use with computer vision, real-time image processing applications and fast training of artificial neural networks (ANN). Open source development environments for computational intelligence and computer vision will be ported to the vector supercomputer.

Multi-computer multi-processor system
Multi-computer systems will be built on the basis of multi-core CPUs (central processing units) and high-speed Ethernet. Possible applications are real-time simulation of rapid dynamic processes, computer vision, training of large artificial neural networks (ANN with more than 1000 neurons) and the evolutionary optimization of the architecture of artificial neural networks for technical applications. The multi-computer multi-processor system with up to 32 CPUs is able to execute professional simulation software on standard multiprocessor capable operation systems.

Artificial Neural Network for the control of assembly-processes in aircraft construction
The behaviour of the material within the side shell of the fuselage section is virtualy modelled with artificial neural networks (ANN). Furthermore a neural-I-controller with the inverse nonlinear characteristic curve of the material behaviour of the side shell is developed to control the assembly process to control the geometry of the side shell at deviation from the requested position back into optimal assembly position.

Optical quality control for soot filters
A non-contact visual system is used for the quality inspection of filter particles. It scans the geometric characteristics of the opening of the filters. Measurements can be replicated on a sub pixel level through standard digital imaging and modern image processing technology. These programs are executed in C or C++ and have short run-times. Quick readings can be conducted with the existing infrastructure.

Energy management in motor vehicles
Vehicle manufacturers are receiving pressure from the EU to manufacture motor vehicles which produce less CO2 emissions. A cars electrical system without a Soft-Hybrid motor would need an average of 600 Watts. Fuel consumption and with it CO2 emission could be reduced by 5% if energy was fed into the electrical system from the energy used in braking, solar energy from solar panels on the roof (photovoltaic) and/or thermoelectric energy from a thermoelectric generator (which has been used in space flight for decades). A sophisticated energy management system based on computational intelligence and microelectronics can optimize energy consumption and reduce CO2 emissions effectively.

Using photogrammetry for measuring free form surfaces and outlines
A photogrammetric system uses two cameras to determine the height of any object with great detail on a value basis. The principles of the system are based on the stereopsis process of two human eyes. A structured surface is essential for an optimum correspondence analysis. Therefore a beamer lights up the structure providing an optimum measurement of faint objects. Combined with automation technologies like robotics, complex processes like lacquering a car can be automated.

Computational Intelligence Controller (CI-Controller), for nonlinear and time-variant control paths
Industrial processes are often non-linear and very complex regarding modelling. Robust rules with optimum quality control and detailed specifications can be too much for even popular procedures. A general approach was developed in this project for a self optimising neuro-adaptive PID industry controller. This controller uses evolutionary optimisation to adapt itself onto an unknown operating point the firm. It then uses parameter-logging (route variables and their associated controller parameters are stored in a database) and an artificial network (ANN) to very quickly recognize a non-linear process in the operating point, and regulate the changing process in accordance with the quality of control requirements. The artificial neural network is trained by the parameters from evolutionary optimisation.  In a continuous operating time, tests show that the regulator behaves like an expert and enhances quality control.   The specifications of the quality criterion and the adaptation come in time periods.
The evolutionary optimisation of the control process model runs parallel with real time control. The control process model is created from measurements of process values from the CI-controller.
More information

Inspecting semi-conductors with a high-definition photon tunnel microscope (PTM)

  

A method was created which increases the resolution of the classic reflected-light microscope, through a prototype named the Photon Tunnel Microscope (PTM) or Simultaneous Optical Near-field Microscope (SONM), which efficiently stimulates the near field. This was put into effect through several diploma thesises [1-11], one dissertation [12] and the collaboration with the firm Leica Microsystems.
A modified conventional reflected-light microscope was used. The microscope was equipped with specially constructed components, which meant the microscopes infrastructure was kept and a cost-efficient modification was possible. More......

Identifying star-shaped outlines on a low contrast picture with genetic algorithms
Evolutionary algorithms are used to solve optimization problems. These algorithms are used in the areas of pattern recognition, structuring of data and especially in areas where classic problem-solving methods fails. The recognition of outlines in a black and white picture is one of the fields of application. The principles of (biological) evolution are consciously imitated. Nature adapts itself to the change of the environment and biological processes are cause of this, these algorithms use these same mechanisms to align onto a concrete black and white picture. Outlines are approximated with B-Spline.
And important field of application for image processing is the field of medical information technology. Image processing is especially used in areas of microscopy and the automatic analysis of medical imagery from radiology. In this type of area, problems arise in the recognition of outlines and the classifying of objects in a black and white picture. Standard outline-detectors will not be able to generate an outline if a picture is blurry or has insufficient light. An approach to solve this problem can be found in an evolutionary algorithm for the recognition of outlines. This was used on artificial, microscopic and medical image data, the results were very good.

Spam filter based on Computational Intelligence for filtering out irrelevant e-mails
Spam mail, also known as unwanted advertisement-emails, is a major problem for the internet. According to a survey by the Radicati Group, 94% of all users say that spam is a big problem for them. But it is not just the users that have a problem. 
Figure 1 shows the growth of costs through spam measured from a company with 10,000 employees. The costs are a result from additional servers needed for processing the spam e-mails.


Figure 1. Costs from spam per mailbox/year. (Spam cost – Radicati Group, California, Palo Alto)

Secure home banking with Computational Intelligence Profile-building for account holders
Security was a priority issue when developing the program: "Mon€yP€nny", a linux based program. The program is started directly from a CD which provides an extra safeguard against sabotage programs simply because the program cannot be rewritten. Mon€yP€nny also uses the latest HBCI standard (Home Banking Computer Interface) when sending data which is cryptographically encrypted. The HBCI standard from 2005 FinTS (Financial Transaction services) is supported in Germany by the major banks, Volksbank, Sparkasse and many others.

Intelligent optimization of storage inventory in businesses
An inhomogeneous logistics process of an international manufacturer went through an analysis using SAP R/3 which analyzed its strengths and weaknesses. Detailed concepts were developed and taught; constructed from a systematic structuring of articles and an inventory-potential analysis, these concepts where created for the optimisation and standardisation of the dispositive logistic functions. Functions for checking for availability in the replace time of a delivery schedule process were optimized, particularly for differentiated consumption control, this was because of a system-level and live implementation of SAP R/3.
Specific criteria’s are automatically assigned to articles from availability-indicators in SAP in an optimum way.

Optimising the parameters for the vagus nerve stimulation with CI-Methods for treating refractory epilepsy in children (OptiVaNeS)
Refractory epilepsy in children cannot be treated with medication and can be very vigorous. The strain on the young patients and the affected families can be alleviated with help from a vagus nerve stimulator (VNS). Such a stimulator is connected to the vagus nerve (VN), this electrode continuously sends electric signals down the nerve stopping important parts of the nerve. The device is comparable to a pacemaker. The vagus nerve is made up of different fibres, each with its own function. That is why through changing the stimulation parameters, the stimulator can create different effects.
The statistical effectiveness of certain parameters was investigated using data mining based on CI-methods within the OptiVaNeS project. Constantly implemented, these parameters can lead to epileptic seizures being suppressed interminably. Numerous adjustment combinations like pulse strength, pulse duration, pulse-width, pause and application duration where reduced from doctors to a small number of effective parameters.

An intelligent system for producing individual skin care products
An intelligent system for the production of individual care products. How does it work? It is base on a computer-supported skin analyser (analysing the fat content, moisture, pH value etc.). The operating program calculates ideal ingredients for the current user’s skin condition. The computer which is based on the computational intelligence method (multicriterial optimisation) investigates a skin care recipe from over 2500 different formulas and the cream for the client is produced directly on site.
The conclusion of the project partner: The skin receives the nutrients it needs for its current state at any time. It does not get more unique! At any time, for any condition, the optimum product for optimum care. A laboratory sample of the system was shown at the exhibition “Medica 2008”.

Remore lectures with Teleteaching (E-Teaching) for further training
Teleteaching has big potential in the future. It belongs to the field of E-learning. Courses or learn events can be spread out over great distances. Presentations (Power Point and alike) along with video and audio are transmitted through data channels, practically without loss of quality. It is possible to receive remote queries on a switchable reverse channel from distant auditoriums.
A lot of areas in life are being influenced by the expanding WWW (World Wide Web). That is why there is a rapid development of so called E-Technologies. The most popular for sure is E-Mail. It does not matter I currently am, it does not matter where I am in the world, if I want to communicate, weather it is with a college in the same building as me who is unfortunately not in his office, or with a university in California, with a college in a working team which is spread out through Europe, with a student who is currently employed abroad somewhere in Europe, it is not a problem sending them a message, important information, delivering research results or working together on a book project. E-Mail has not only enriched, accelerated and improved foreign mail but also the area of internal communication. E-Mail programs have become E-Registries, which are as effective as the registries in a ministry. An example of this is Microsoft Outlook.

E-Learning for further training in the field of automation/processing informatics
Different types of e-Learning tools are used in the South Westphalia University of Applied Sciences (SWUAS) in Iserlohn in the faculty Computer Science and Natural Science (I+N).
The scope from teleteaching is enough for e-learning which is based on open source platforms, professional platforms to blended learning. An e-learning/e-simulation approach was developed which included conceptions and implementations from java applets. This was developed for students who have an occupation parallel to their bachelor (BA) or master (MA) study course who can develop their skills on simulation systems without having a deep understanding of it, even if their field of knowledge is analysis, modelling and simulation of process-informatics and cybernetics. An evaluation of the e-learning tools leads to a good and as far as an excellent result. An additional evaluation showed that the open source e-Learning platform Moodle received good results when in connection with java applets for process-informatics and cybernetics.