Developing an HTN Planner for Virtual Ecosystems
Developed decades ago, Hierarchical Task Network Planners are now starting to be used in the video game industry as part of their AI systems. Due to its expressiveness and efficiency with complex models, they are a good alternative to the widely used State Machines or Behavior Trees. Despite the gre...
| Author: | |
|---|---|
| Format: | master thesis |
| Publication Date: | 2023 |
| Country: | España |
| Institution: | Universitat Politècnica de Catalunya (UPC) |
| Repository: | UPCommons. Portal del coneixement obert de la UPC |
| Language: | English |
| OAI Identifier: | oai:upcommons.upc.edu:2117/399252 |
| Online Access: | https://hdl.handle.net/2117/399252 |
| Access Level: | Open access |
| Keyword: | Video games Artificial intelligence Hierarchical Task Network (HTN) Planning Unreal Engine 5 Intel·ligència artificial IA per Videojocs C/C++. Artificial Intelligence AI for Video Games Videojocs Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial |
| Summary: | Developed decades ago, Hierarchical Task Network Planners are now starting to be used in the video game industry as part of their AI systems. Due to its expressiveness and efficiency with complex models, they are a good alternative to the widely used State Machines or Behavior Trees. Despite the great opportunities that this technique offers, there's a lack of examples and optimizations for video games. In this project we will develop an HTN Planning system, including the Planner algorithm, domain language and the entities' domain, to recreate a Virtual Ecosystem with a concrete thematic. Our HTN Planning system will be based on the SHOP2 system from Dana Nau. It will be implemented in C/C++ and executed using the video game engine "Unreal Engine 5", while trying to optimize it for the chosen ecosystem. |
|---|