The
CLUE project is on the topic of AI in computer game and offers a convenient
simulation environment for intelligent methods exploration for surveillance
systems. As is known, CLUE is a poplar game and its object is to find the
answer to the hidden three cards, i.e. who, where and how. The key issue in
CLUE for a player is suggestion making including the sub-problem of optimal
movement. The motivation of research on Clue is based on the similarity of
surveillance systems such as landmine detection and Clue: suggestion making
<-> sensor measurement; optimal movement <-> platform (robot)
path planning. Many interesting problems are involved in the game
simulation, such as, probabilistic inference, decision making, preposterior decision analysis, path planning, etc. A
BN approach based on updated evidence table (ET) including both hard and
soft evidence is proposed to solve the problem of computer player's
intelligent suggestion making in the computer game of CLUE. CLUE belongs to
a class of problem, so-called Treasure
Hunt Problem.
Conducting
the research at Duke University
is Dr.
Silvia Ferrari , Assistant Professor in
Mechanical Engineering at Duke University
, and her graduate student, Chenghui Cai.
CLUE®
game board. CLUE® & ©2006 Hasbro, Inc. Used with permission.
