Summaries - Office of Research & Innovation
Back Mission Command Analysis Using Monte Carlo Tree Search, Part II
|Division||Research & Sponsored Programs|
|Department||Modeling, Virtual Environments & Simuation Institute|
|Investigator(s)||Darken, Christian J.|
|Sponsor||Army TRADOC Analysis Center-Monterey (Army)|
TRAC-Monterey is carrying out supporting research to produce a documented and tested methodology that applies Monte Carlo Tree Search methods to decision situations in order to expand mission command oriented analysis. Mission command features decentralized execution with subordinate commanders exercising disciplined initiative while acting aggressively and independently to accomplish the mission within the commander's intent. The methodology will improve analysis by extending data developed from operational data, wargames, and other subject-matter expertise elicitation into a simulation environment where more extensive and rigorous analysis can be accomplished.
An outstanding challenge of modeling mission command is in modeling the situation awareness of the combatants involved. This is especially true when attempting to apply a search algorithm like Monte Carlo Tree Search which originated as a technique for modeling games of perfect information. In order to apply this type of algorithm in a domain where uncertainty is present, assumptions need to be made about the state of the battlefield, and these assumptions need to be consistent with the situation awareness model of the simulated commander.
|Publications||Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal|
|Data||Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal|