The Thursday Theory post this week takes another whitepaper from from the Symposium on Interactive 3D Games and Graphics 2007 that’s helpful for game AI, and breaks it down into the basic ideas. See the previous post in this series for more about character animation.
The combination of motion capture playback with motion synthesis by a planner into a hybrid animation system,
A way to control the resulting motion using different sampling strategies for the randomized planner.
Such a solution provides the realism of motion capture for full-body locomotion, but at the same time allows the animator (or runtime logic) to constrain and retarget parts of the body animation to be physically accurate (e.g. the arm or leg).
View or download the movie (MP4, 39 Mb).
Here’s the abstract:
Editing recorded motions to make them suitable for different sets of environmental constraints is a general and difficult open problem. In this paper we solve a significant part of this problem by modifying full-body motions with an interactive randomized motion planner. Our method is able to synthesize collision-free motions for specified linkages of multiple animated characters in synchrony with the characters’ full-body motions.
The proposed method runs at interactive speed for dynamic environments of realistic complexity. We demonstrate the effectiveness of our interactive motion editing approach with two important applications: (a) motion correction (to remove collisions) and (b) synthesis of realistic object manipulation sequences on top of locomotion.
Download the paper from the site (PDF, 5.4 Mb):
Interactive Motion Correction and Object Manipulation A. Shapiro, M. Kallmann, P. Faloutsos Proceedings of Symposium on Interactive 3D Graphics and Games 2007
Here’s a short assessment of the technology based on how simple it would be to use in games.
- Applicability to games: 7/10
- The number of games that require fully accurate motion planning for limbs is very small. It doesn’t really matter in a third-person shooter if someone’s foot goes through a small object on the floor. That said, with the push towards next-generation animation, this kind of attention to detail is required.
- Usefulness for character AI: 7/10
- Making actors interact realistically with their environment is always a good thing, and this technology allows them to manipulate objects while moving. Again, it’s not a requirement, but it’s a nice touch.
- Simplicity to implement: 6/10
- This algorithm is surprisingly easy; the planner samples random points in space towards the goal, builds a path out of it, then smooths it out. That trajectory is then used by a traditional IK algorithm, which would take a bit more effort to implement.
Screenshot: An umbrella avoiding two posts.
What are your thoughts on applying such technology to game AI?