Way to go: A framework for multi-level planning in games Norman Jaklin Wouter van Toll Roland Geraerts Department of Information and Computing Sciences Utrecht University, The Netherlands
Way to go: A framework for multi-level planning in games Norman Jaklin Wouter van Toll Roland Geraerts Department of Information and Computing Sciences.
Dec 17, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Way to go: A framework for multi-level planning in games
Norman Jaklin Wouter van Toll Roland Geraerts
Department of Information and Computing Sciences
Utrecht University, The Netherlands
Way to go: A framework for multi-level planning in games 2
About us
Norman Jaklin, Wouter van Toll: PhD Students
Roland Geraerts: Assistant Professor
Part of the Games and Virtual Worlds groupResearch on animation, motion planning, ...
For us: path planning and crowd simulation
10-6-2013
Way to go: A framework for multi-level planning in games 3
Path planning in games
10-6-2013
Finding a path for one characterA* on a grid
Navigation meshes
Many characters: crowd simulationIndividual goals
Collision avoidance
Problem solved?Approximations, assumptions, memory
We need an efficient and flexible framework
Geraerts, R.J. (Roland)
and extendible
Geraerts, R.J. (Roland)
you could add a screenshot of advanced path planning in games, e.g. Starcraft; otherwise, we're too negative
Geraerts, R.J. (Roland)
show the video instead of the picture
Way to go: A framework for multi-level planning in games 4
Planning hierarchy
Five levels
5: ICAPS / PG
4/3/2: our frameworkNavigation mesh
Indicative routes
Velocity-based models
10-6-2013
Way to go: A framework for multi-level planning in games 5
Our framework
Explicit Corridor Map: navigation meshComputed automatically
Clearance info: supports all character sizes
2D and multi-layered environments
Dynamic updates
Visibility queries
10-6-2013
Way to go: A framework for multi-level planning in games 6
Our framework
Level 4: Route planningMedial axis corridor
Any route in the corridor (e.g. the shortest)
Or: visibility, crowd density, ...
Result: indicative route (IR)
10-6-2013
Way to go: A framework for multi-level planning in games 7
Our framework
Level 3: Route followingCharacter follows IR smoothly: attraction point
Indicative Route Method: basic
MIRAN: supports terrain preferences
Level 2: Local movementPreferred velocity actual velocity
Collision avoidance: many models exist
10-6-2013
Way to go: A framework for multi-level planning in games 8
Demo
10-6-2013
Way to go: A framework for multi-level planning in games 9
Current / Future work
Crowds in dynamic environments
Combinations with... Game engines (Unity)
High-level planning?
10-6-2013
Related Documents
