|
|
Reducing Complexity of AI in Open-World Games by Combining Search-based and Reactive Techniques
Černý, Martin ; Brom, Cyril (advisor) ; Dignum, Frank (referee) ; Pilát, Martin (referee)
Open-world computer games present the players with a large degree of freedom to interact with the virtual environment. The increased player freedom makes open-world games a challenging domain for artificial intelligence. In this thesis we present three novel techniques to handle various types of complexity inherent in developing artificial intelligence for open-world games. We developed behavior objects that extend the well-known concept of smart objects and help in structuring codebase for reactive reasoning, we propose and implement constraint satisfaction techniques to specify behavior from a global viewpoint and we have shown how adversarial search techniques can mitigate the need for complex reactive decision mechanisms when a large number of parameters has to be taken into account. The general techniques are implemented and evaluated in the context of a complete open-world game Kingdom Come: Deliverance. Powered by TCPDF (www.tcpdf.org)
|
|
|
Implementation of the SF-HRP action selection mechanism
Farka, František ; Plch, Tomáš (advisor) ; Dvořák, Filip (referee)
In this thesis, we present our C++ implementation of the State-Full Hierarchical Reactive Planning (SF-HRP) mechanism for action selection for virtual agents. The implementation is connected to 3D virtual environment and provides access to 3rd party software for profiling purposes vie defined interface. A prototype of such a profiler is part of the implementation. The thesis also presents an input format for agent's behavior description and is used within the implementation. Both the implementation and input format are demonstrated on testing scenarios. The SF-HRP concept is discussed with respect to the difficulty of design of agent's behavior and complexity of the implementation.
|
|
|
Reducing Complexity of AI in Open-World Games by Combining Search-based and Reactive Techniques
Černý, Martin ; Brom, Cyril (advisor) ; Dignum, Frank (referee) ; Pilát, Martin (referee)
Open-world computer games present the players with a large degree of freedom to interact with the virtual environment. The increased player freedom makes open-world games a challenging domain for artificial intelligence. In this thesis we present three novel techniques to handle various types of complexity inherent in developing artificial intelligence for open-world games. We developed behavior objects that extend the well-known concept of smart objects and help in structuring codebase for reactive reasoning, we propose and implement constraint satisfaction techniques to specify behavior from a global viewpoint and we have shown how adversarial search techniques can mitigate the need for complex reactive decision mechanisms when a large number of parameters has to be taken into account. The general techniques are implemented and evaluated in the context of a complete open-world game Kingdom Come: Deliverance. Powered by TCPDF (www.tcpdf.org)
|
|
|
Implementation of the SF-HRP action selection mechanism
Farka, František ; Plch, Tomáš (advisor) ; Dvořák, Filip (referee)
In this thesis, we present our C++ implementation of the State-Full Hierarchical Reactive Planning (SF-HRP) mechanism for action selection for virtual agents. The implementation is connected to 3D virtual environment and provides access to 3rd party software for profiling purposes vie defined interface. A prototype of such a profiler is part of the implementation. The thesis also presents an input format for agent's behavior description and is used within the implementation. Both the implementation and input format are demonstrated on testing scenarios. The SF-HRP concept is discussed with respect to the difficulty of design of agent's behavior and complexity of the implementation.
|
guest ::
