The nonlinear control techniques are applied to the model of rotary inverted pendulum. The model has two degrees of freedom and is not exactly linearizable. The goal is to control output trajectory of the rotary inverted pendulum asymptotically along a desired reference. Moreover, the designed controller should be robust with respect to specified perturbations and parameters uncertainties. A combination of techniques based on nonlinear normal forms, output regulation and sliding mode approach is...

Traditional data mining methods based on rough set theory focus on extracting models which are good at classifying unseen obj-ects. If one wants to uncover new knowledge from the data, the model must have a high descriptive quality-it must describe the data set in a clear and concise manner, without sacrificing classification performance. Rough modeling, introduced by Kowalczyk (1998), is an approach which aims at providing models with good predictive emphand descriptive qualities, in addition to...

A computationally simple method for generating reduced-order models that minimise the $L_2$ norm of the approximation error while preserving a number of second-order information indices as well as the steady-state value of the step response, is presented. The method exploits the energy-conservation property peculiar to the Routh reduction method and the interpolation property of the $L_2$-optimal approximation. Two examples taken from the relevant literature show that the suggested techniques may lead to...

In this paper, a new concept of the Reactivity Trace Curve (RTC) for reactor power control is presented. The concept is demonstrated for a reactor model with one group of delayed neutrons, where the reactivity trace curve is simply a closed form exponential solution of the RTC-differential equation identifier. An extended reactor model of multigroup (six groups) of delayed neutrons is discussed for power control using the RTC-method which is based on numerical solution of the governing equation...

The paper discusses the problem of rule weight tuning in neuro-fuzzy systems with parameterized consequences in which rule weights and the activation of the rules are not interchangeable. Some heuristic methods of rule weight computation in neuro-fuzzy systems with a hierarchical input domain partition and parameterized consequences are proposed. Several heuristics with experimental results showing the advantage of their usage are presented.