Theory of dynamic advising has been converted into universal tool based on dynamic normal mixtures used as environment model and on fully probabilistic design used as constructor of advising strategies. The resulting system has been applied in such diverse field as operating of rolling mills and supporting of medical doctors curing cancer of thyroid gland. Other applications are addressed, too.
The only way to solve the problem of bad traffic situation in cities is the application of signal traffic control. Daily traffic course and the requirements of public transport preferences strongly need the application of some intelligent traffic system. Such a system should provide some efficient traffic control using on-line adaptive methods.
Complex physical modeling, need for fast on-line computations on a large modeled time-spatial domain and sparse data are dominant characteristics of this important applied project.
These applications while being important on their own help us to deal with problems relying on prior information, see
lymphoscintigraphy, and careful modeling, see
nuclear. All of them including
internal test modeling capabilities of mixtures especially in cases when the number of learning data is small.
Control of electric drives is a well studied area and many techniques are available for normal opration regimes. However, even the best the state of the art methods fail under critical operating conditions such as broken sensors or extremely low speed. Application of Bayesian decision making algorithms helps to improve this situation and provide more reliable and robust control strategies.
Solving common projects is based on effort to utilize knowledge and experience academic/scientific potential ÚTIA AV ČR, v.v.i. and experimental background of car maker
ŠKODA AUTO a.s. The aim is not only to set up knowledge base for research projects like
EKODRIVE and
DAR but also to specificate shared outputs in the form of vehicle units, simulation tools and mathematical models as the source for further development advanced assistant driver systems.
One of the key objectives of any rolling mill control system is to keep the thickness of the processed material within the prescribed tolerance band, which can be as low as +-10 micrometers for thin strips. Failure to comply with the tolerances results in losses which, according to experts estimate, might go up to 10% of the profit for poorly equipped rolling mills.