We are a group of students from the Faculty of Computer Science at the Complutense University of Madrid (Spain). We have developed, as part of our Final Degree Project supervised by Federico Peinado (Narratech Laboratories), a code plugin for Unreal Engine 4 called GOAP NPC. This plugin aims to help with the creation of Artificial Intelligence (AI) of Non-Player Characters (NPCs), using a Goal-Oriented Action Planning (GOAP) architecture.
Most developers usually use Behavior Trees (BTs) to define what NPCs can do. It seems intuitive and easy to get started with. However, as the complexity of the scenario grows, it becomes really hard to handle. That's when GOAP comes into play. One of the most important differences with BTs is the ability of GOAP to react to external changes and dynamically modify the character's plan during runtime. You can obtain more information about GOAP here: http://alumni.media.mit.edu/~jorkin/GOAP_draft_AIWisdom2_2003.pdf
Currently we are contacting developers who have a minimum of knowledge about developing videogames with Blueprints on Unreal Engine to ask for help in testing our tool and sharing their experience. The test involves the creation of a simple controller for an NPC with a few actions and both kind of approaches (firstly using BTs, and then using GOAP).
We greatly appreciate the help that you can provide us. It will be a very useful contribution to the improvement of our plugin, and will help everyone to better understand the usefulness of AI planning tools. If you have any questions or want to know more about it, please contact us by email: email@example.com
Thank you very much!
- Diego Romero, Mario Sánchez & José Manuel Sierra
Please, DOWNLOAD the GOAP NPC Code Plugin from the Unreal Marketplace, and INSTALL it in the last version of the UE4 engine: https://www.unrealengine.com/marketplace/en-US/product/goap-npc-goal-oriented-action-planning-for-non-player-characters
Now you are ready to start the test, continue to the next section.