|
|
Study and comparison of main kinds of evolutionary algorithms
Štefan, Martin ; Holeňa, Martin (advisor) ; Gemrot, Jakub (referee)
Evolutionary algorithms belongs among the youngest and the most progressive methods of solving difficult optimization tasks. They received huge popularity mainly due to good experimental results in optimization, a simplicity of the implementation and a high modularity, which is an ability to be modified for different problems. Among the most frequently used Evolutionary algorithms belongs Genetic Algorithm, Differential Evolution and Evolutionary Strategy. It is able to apply these algorithms and theirs variants to both continuous, discrete and mixed optimization tasks. A subject of this theses is to compare three main types of algorithms on the catalyst optimization task with mixed variables, linear constraints and experimentally evaluated fitness function.
|
|
|
Triplanetary
Huječek, Adam ; Cibulka, Josef (advisor) ; Gemrot, Jakub (referee)
The main purpose of this work is to bring the almost forgotten turn-based board game Triplanetary from the 70s and the beginning of the 80s of the last century to the screens of today's computers. The program allows multiple players to play on one computer in the so called hotseat mode two of the several scenarios available in the original game - racing Grand Tour with the option to play with computer controlled opponents and battle Nova for three players. However thanks to the suitable design it is easy to implement the rest of scenarios or of course add completely new ones provided the user has the knowledge of JAVA which the game is programmed in. Another advantage is the option to save and exit the game at any time and return to it later.
|
|
|
Comparing reactive techniques to classical planning for intelligent virtual agents
Černý, Martin ; Gemrot, Jakub (advisor) ; Balyo, Tomáš (referee)
Many contemporary computer games can be described as dynamic real-time simulations inhabited by autonomous intelligent virtual agents (IVAs) where most of the environment structure is immutable and navigating through the environment is one of the most common activities. Though controlling the behaviour of such agents seems perfectly suited for action planning techniques, planning is not widely adopted in existing games. This paper attempts to answer the question whether the current academic planning technology is ready for integration to existing games and under which conditions. The paper compares reactive techniques to classical planning in handling the action selection problem for IVAs in game-like environments. Several existing classical planners that occupied top positions in the International Planning Competition were connected to the virtual environment of Unreal Development Kit via the Pogamut platform. Performance of IVAs employing those planners and IVAs with reactive architecture was measured on a class of game like test environments under different levels of external interference. It was shown that agents employing classical planning outperform reactive agents only if the size of the planning problem is small or if the environment changes are either hostile to the agent or not very frequent.
|
|
|
Petrobras Planning Domain: PDDL Modeling and Solving
Hanes, Martin ; Toropila, Daniel (advisor) ; Gemrot, Jakub (referee)
The thesis explores Petrobras domain from the challenge track of the ICKEPS 2012 competition, which is an example of a difficult problem stand- ing on the borderline between planning and scheduling. It describes state-of-art techniques of the modern artificial intelligence that were chosen for this purpose. It guides the reader in the process of modeling the Petrobras domain in PDDL language and explains the results provided by the existing tools used within plan- ning community. In the end it compares the results of the chosen approach to the other approaches used in ICKEPS 2012 competition, showing that the existing sequential planning systems are usable only for solving smaller problem instances of this domain.
|
|
|
Using artificial neural networks to control genetic algorithms
Dörfler, Martin ; Holeňa, Martin (advisor) ; Gemrot, Jakub (referee)
Genetic algorithms are some of the most flexible among optimization methods. Because of their low requirements on input data, they are able to solve a wide array of problems. The flexibility is balanced by their lower effectiveness. When compared to more specialized methods, their results are inferior. This thesis examines the possibility of increasing their effectiveness by means of controlling their run by an artificial neural network. Presented inside are means of controlling a run of a genetic algorithm by a self-organizing map. The thesis contains an algorithm proposal, a prototype implementation of such algorithm and a series of tests to assess its efficiency. While the results on benchmark functions show some positive properties, the problems of greater complexity yield less optimistic results.
|
|
|
Artificial Intelligence for modern card games
Vejmola, Jiří ; Sýkora, Ondřej (advisor) ; Gemrot, Jakub (referee)
In this work we study the ways of implementing arti cial intelligence for modern card games. We strive to create various players, to put them into matches against each other. By doing so we can verify the efficiency of employed methods, such as simulation, random approach, opponent modeling, xed strategy, etc. We would also like to nd out what influence do other game elements have on the progress of the game and the performance of individual players. The game used in the thesis is based on Magic: The Gathering's rules for two players, speci cally on Magic 2010 Core Set, which revised certain parts of the game. These rules are then modi ed or simpli ed when necessary for implementing game mechanics on a computer. Decks of cards used in testing are based on the existing ones.
|
|
|
Moderní evoluční algoritmy pro hledání oblastí s vysokou fitness
Káldy, Martin ; Holeňa, Martin (advisor) ; Gemrot, Jakub (referee)
Evolutionary algorithms are optimization techniques inspired by the actual evolution of biological species. They use conceptually simple process of two repeating phases of reproduction and fitness-based selection, that iteratively evolves each time better solutions. Evolutionary algorithms receive a lot of attention for being able to solve very hard optimization problems, where other optimization techniques might fail due to existence of many local optima. Wide range of different variants of evolutionary algorithms have been proposed. In this thesis, we will focus on the area of Estimation of Distribution Algorithms (EDA). When creating the next generation, EDAs transform the selected high-fitness population into a probability distribution. New generation is obtained by sampling the estimated distribution. We will design and and implement combinations of existing EDAs that will operate in business-specific environment, that can be characterized as tree-like structure of both discrete and continuous variables. Also, additional linear inequality constraints are specified to applicable solutions. Implemented application communicates with provided interfaces, retrieving the problem model specification and storing populations into database. Database is used to assign externally computed fitness values from...
|
|
|
Výukové scénáře v projektu Pogamut
Vansa, Radim ; Gemrot, Jakub (referee) ; Kadlec, Rudolf (advisor)
: The goal of this project is to extend Pogamut platform with a tool for non-programmers, especially teachers, who want to use 3D virtual environment as a learning medium. They should be able to easily write scenarios, where they can test their students in different situations. One part of this bachelor thesis is the implementation of sample scenario proving student's abilities of orientation in urban environment. These scenarios can be scripted in naturally looking, localizable language interpreted in the Drools rule based engine. The Pogamut project allows to present these scenarios in first-class virtual reality worlds, for instance in Unreal Tournament 2004.
|
|
|
Koordinace chování virtuálních lidí
Gemrot, Jakub ; Neruda, Roman (referee) ; Brom, Cyril (advisor)
This thesis is about specific approach to the behavior coordination of multiple embodied virtual agents. Agents may act for themselves or be controlled directly by bodiless coordination agents. This kind of approach is designed for the area of interactive storytelling, where the actor agents are viewed as a string puppets that are controlled by the abstract director. The control mechanism is based upon the BDI architecture and the AgentSpeak(L) language that is extended with template plans and new plan execution mechanism that allows the directing of other actor agents.
|
|
|
Mapping, processing, and representation of a map for Unreal Tournament bots
Gemrot, Jakub ; Brom, Cyril (advisor) ; Yaghob, Jakub (referee)
In many computer games the navigation points are used for the movement of computer-controled players. With the growing power of desktop computers more sophisticated methods are used for the movement of players, namely the method of NavMesh. NavMesh is a set of convex polygons continuously covering the reachable places inside the game by characters. This approach was used in the recently produced games Half-Life 2 and F.E.A.R. Those commercial products can't be used by the academy, therefore the chance to experiment with NavMesh doesn't exist. But there is a simple way to control the characters in environment of older game Unreal Tournament. This work is about the advantages of using NavMesh against older approach of navigation points. This work then presents the way how to automatically generate the NavMesh for the maps of Unreal Tournaments. NavMesh is created using the algorithm from computer graphics for the conversion from CSG representation of the map into boundary representation out of which the NavMesh can be generated. The work also contains a demonstration of usage of NavMesh for the movement of players in Unreal Tournament's maps.
|
guest ::
