/ Tsuneyuki Hiramoto / Professor
/ Lothar M. Schmitt / Associate Professor
/ Jens Herder / Research Associate
Application-oriented, the Computer Industry Laboratory tries to enhance production and engineering processes in industry. A deep understanding of the art of working is required.
Research at the university and in industry has to be brought together to achieve advances for humanity. To coordinate industrial development, to provide a basis for further products, and to save financial investment, it is necessary to introduce standards. The Computer Industry Laboratory would like to influence the standardization process in new areas and open it for future needs, not to re-establish existing systems.
Currently, Mr. Herder participates in the Intelligent Dental Care System Project where he manages the design of the user interface. The research of Professor Hiramoto is reliability, especially applied to the safety standards of nuclear power plants. Professor Schmitt participates in research on mathematical models for genetic algorithms used for chip placement problems. Furthermore, he participates in research on the modeling of semiconductor devices with computer algebra methods.
With a top-down education approach, students are involved in joint research projects with industry. They learn engineering by doing to it. This increases their motivation too.
Current Research Topics:
CAD/CAM techniques are used increasingly in dentistry for design and fabrication of teeth restorations. An important issue is preserving occlusal contacts of teeth after restoration. Traditional techniques based on the use of casts with mechanical articulators require manual adjustment of occlusal surface, which becomes impractical when hard restoration materials like porcelain are used; they are also time and labor consuming. Most existing computer systems ignore completely such an articulation check, or perform the check at the level of a tooth and its immediate neighbors. We present a new mathematical model and a related user interface for global articulation simulation, developed for the Intelligent Dental Care System project. The aim of the simulation is elimination of the use of mechanical articulators and manual adjustment in the process of designing dental restorations and articulation diagnostic. The mathematical model is based upon differential topological modeling of the jaws considered as a mechanical system. The user interface exploits metaphors that are familiar to dentists from everyday practice. A new input device designed specifically for use with articulation simulation is proposed.
A mathematical model of genetic algorithms is developed. First, an analysis of the mutation and crossover operations was done which is based upon linear algebra and spectral theory for selfadjoint matrices. This is pretty much completed. It leads to valuable insight into the mechanism of the algorithm. In particular, mutation and crossover are better understood as contracting maps on the (probability-) state space. And, it has become mathematically clear why crossover accelerates the algorithm. A conjecture by Vose, one of the experts in the field, can be proven. At the moment, we are studying application of a generalized version of a theorem due Hajek who gave a cooling schedule for simulated annealing.
We use a genetic algorithm to adjust graphics data and other data obtained in different measurements.