View, browse and sort the ever-growing list of sessions by day, time, pass type, topic, and format. With this Session Viewer, you can view session and speaker details for Game Developers Conference 2024.You will be able to build your schedule with the GDC Mobile App. The GDC 2024 app will be available for download in Apple Apps and Google Play late February 2024.
Sessions do fill up and seating is first come, first serve, so arrive early to sessions that you would like to attend. Adding a session to your schedule does not guarantee you a seat.
Lingxiao Fu (Advanced AI Programmer, Rocksteady Studios)
Location: Room 2005, West Hall
Date: Friday, March 22
Time: 1:30 pm - 2:30 pm
Pass Type: All Access Pass, Core Pass - Get your pass now!
Vault Recording: TBD
Audience Level: Intermediate
This talk presents an expressive and scalable AI movement stack that realizes diverse AI traversal behaviors in Suicide Squad: Kill The Justice League. A hierarchical multi-modal pathfinding approach operating on hyper-connected navigation data is used to produce effective paths in the complex open world of Metropolis. A novel path-planning approach is introduced between pathfinding and path following to create more varied and nuanced traversal decisions. AI path is annotated and broken down into featured patterns based on agent movement constraints and preference and then solved by an HTN-style planner into traversal plans. An adaptive traversal plan execution layer allows AI agents to carry out the planned traversal with the ability to repair and adjust their actions on the fly.
The audience will learn how to construct a movement stack that offers more flexibility and scalability than present movement pipelines. They will learn how HTN planning is adapted to solve an AI traversal planning problem. This will empower their agents to traverse in various ways.
This is for those interested in how to realize diverse AI traversal in complex world representations, based on how the AI movement stack of Suicide Squad: Kill The Justice League operates. Prerequisite knowledge of pathfinding, navigation meshes, and automated planning is useful.