Differences
This shows you the differences between two versions of the page.
| — |
theme:planning_in_real-time_domain [2013/02/18 16:18] (current) martin.cerny created |
||
|---|---|---|---|
| Line 1: | Line 1: | ||
| + | < | ||
| + | < | ||
| + | <td width=" | ||
| + | <td> </ | ||
| + | < | ||
| + | </ | ||
| + | < | ||
| + | <td width=" | ||
| + | <td> </ | ||
| + | < | ||
| + | </ | ||
| + | < | ||
| + | <td style=" | ||
| + | <td> </ | ||
| + | < | ||
| + | </ | ||
| + | < | ||
| + | <td style=" | ||
| + | <td> </ | ||
| + | < | ||
| + | <p> | ||
| + | AI planning techniques are a well studied area, but still, planning is not widely used to control agents in computer games and other real-time environments. With the recent increase in computing power, the only remaining issue seems to be the relative lack of information flow between the planning and virtual agents scientific communities. | ||
| + | </ | ||
| + | In our recent work we experimented with employing AI planning in real-time decision making of a virtual agent in UDK (Unreal developoment kit) with Pogamut platform. Your work will be to design another class of dynamic environments that require the agent to plan ahead. Then you will compare performance of agents that employ several existing AI planners with results of agents with simple reactive architecture. | ||
| + | </ | ||
| + | <p> | ||
| + | Since one similar experiment was already performed in Pogamut, most of the " | ||
| + | </ | ||
| + | <p> | ||
| + | Perfect for anybody who wants to enter the domain of general planners! | ||
| + | </ | ||
| + | </ | ||
| + | </ | ||
| + | </ | ||