No abstract

No abstract

With the increasing availability of robots and image processing systems, the automated robotic line implementations are still extensively increasing. As well, the complexity of these systems increases, which increases the demands on the determination of their function and possible analysis. For these purposes the complex robotic systems are modeled. Here, the Coloured Petri Nets were used to model the robotic sorting line. Specifically, the CPN Tools software was used to model creation, simulation, analysis and to made experiments. State space analysis was done to revelation of possible unwanted deadlocks. Furthermore, the sorting line model was used to found out the maximum number of objects that can be sorted in a defined time. As the robot manipulation time directly affects maximum number of objects in one batch, the timing procedure was created to declare robot operation dependency on types of manipulated object. The state space analysis showed that there are no unwanted deadlocks in the system. The model has been verified and declared as correct. Optimal composition and order of objects was successfully found by timed coloured Petri nets experiments.

No abstract

The contribution is focused on helicopter elevation control. The experiments are performed on helicopter model, which is significantly nonlinear plant and it simulates in some simplified way dynamics of real helicopter. There is derived new way to control it which uses piecewise-linear neural model. As it is shown at the end of the paper, this new control algorithm brings decent performance improvement compared to certain classical control technique.