Thesis
Improving the survivability of agents in a first-person shooter urban combat simulation by incorporating military skills
Washington State University
Master of Science (MS), Washington State University
2007
Handle:
https://hdl.handle.net/2376/103298
Abstract
The exhibition of intelligence while selecting paths in a combat setting should be based upon the right balance of the risk involved and the traversal time. In this work we propose an algorithm that finds strategic paths inside an urban combat game map with a set of enemies. The strategic path calculation is based upon the hit probability calculated for each enemy's weapons and the risk vs. time preference. Ultimately, the strategic path calculation minimizes both time and risk as per mission objectives. The strategic path planning concept can be applied to both Real Time Strategy (RTS) and First Person Shooter (FPS) games. We propose evaluating a map at two levels of abstraction: Area level and Grid level. Area level strategic path computation can be done at run-time, because Areas are far less in count compared to Waypoints. When the agent reaches the computed Area, the strategic path is computed over the Grid Points of that Area. Thus, the calculation of the hit probability can take into account the real-time movements of the enemies as the agent traverses the Grid Points of an Area. Secondly, in addition to the computational savings of calculating strategic paths at the Area level, rather than the Grid level (or using Waypoints), there is also the issue of not knowing visibility details within an Area until the agent arrives at that Area, especially in urban combat settings. Thus the agent exhibits intelligence (in strategic path computation) in a more realistic way. We performed out-game and in-game experiments on our proposed model of strategic path computation and found that the computed strategic paths based upon a high risk vs. time preference are significantly safer.
Metrics
4 File views/ downloads
11 Record Views
Details
- Title
- Improving the survivability of agents in a first-person shooter urban combat simulation by incorporating military skills
- Creators
- Ashish C. Singh
- Contributors
- Lawrence B. Holder (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; Pullman, Wash. :
- Identifiers
- 99900525048001842
- Language
- English
- Resource Type
- Thesis