|
|
New Cellular Automata Design Techniques
Baláž, Martin ; Drábek, Vladimír (referee) ; Bidlo, Michal (advisor)
The aim of this master thesis is to introduce a new technique for the design of cellular automata which will provide a better possibilities for the implementation and solving given problems in an environment of non-uniform automata. In this work, the theoretical foundations of cellular automata have been summarized and the possibilities of their design were examined using two evolutionary principles that have commonly been used - genetic algorithm and cellular programming. Two principally different issues were selected on which the possibilities and capabilities of these techniques were proven: the synchronization problem and the system of implementation of logic gates in an environment of cellular automata. Based on a review of the implementation properties and the initial results of usage of these methods a new design method for cellular automata was created - cellular evolution. The cellular evolution with its method of "prediction of the future state of surrounding cells" provides new possibilities in the design of cellular automata since it operates with structured genes which allow the gene to be active for a variety of cellular surroundings. In the conclusion of this work, all three methods were compared on two selected problems and their abilities were summarized in a detailed overview.
|
|
|
Cellular Automaton in Dynamical Environment
Bendl, Jaroslav ; Jaroš, Jiří (referee) ; Bidlo, Michal (advisor)
This bachelor thesis focuses on the method of evolution of cellular automaton capable of self-repair after being damaged by external environment. The described method is based on cellular programming algorithm and uses principles of biological development. Experiments leading to verification of regenerative ability for cellular automaton evolved by this approach are presented in this work.
|
|
|
New Cellular Automata Design Techniques
Baláž, Martin ; Drábek, Vladimír (referee) ; Bidlo, Michal (advisor)
The aim of this master thesis is to introduce a new technique for the design of cellular automata which will provide a better possibilities for the implementation and solving given problems in an environment of non-uniform automata. In this work, the theoretical foundations of cellular automata have been summarized and the possibilities of their design were examined using two evolutionary principles that have commonly been used - genetic algorithm and cellular programming. Two principally different issues were selected on which the possibilities and capabilities of these techniques were proven: the synchronization problem and the system of implementation of logic gates in an environment of cellular automata. Based on a review of the implementation properties and the initial results of usage of these methods a new design method for cellular automata was created - cellular evolution. The cellular evolution with its method of "prediction of the future state of surrounding cells" provides new possibilities in the design of cellular automata since it operates with structured genes which allow the gene to be active for a variety of cellular surroundings. In the conclusion of this work, all three methods were compared on two selected problems and their abilities were summarized in a detailed overview.
|
|
|
Cellular Automaton in Dynamical Environment
Bendl, Jaroslav ; Jaroš, Jiří (referee) ; Bidlo, Michal (advisor)
This bachelor thesis focuses on the method of evolution of cellular automaton capable of self-repair after being damaged by external environment. The described method is based on cellular programming algorithm and uses principles of biological development. Experiments leading to verification of regenerative ability for cellular automaton evolved by this approach are presented in this work.
|
guest ::
