Title :
A New Plan Recognition Algorithm and its Implementation
Author :
Dan Zhang ; Longmei Jie ; Guoqiang Shao
Author_Institution :
Comput. Sci. & Inf. Technol. Dept., DaQing Normal Univ., DaQing, China
Abstract :
In order to efficiently recognize the partially ordered or interleaved, multiple plans, treat the partial observability of the domains and exclude misleading actions, in this paper, we designed a new plan recognition algorithm based on extending and pruning the Explanation Graph and then implemented it using the VC++ programming language. The results of test experiment indicate that the algorithm we presented here can completely accomplish the expected effect. It can not only recognize the plan and goal of the agent being performed efficiently, but also predict the actions the agent will take, and can also exclude the misleading actions.
Keywords :
C++ language; graph theory; multi-agent systems; pattern classification; planning (artificial intelligence); programming languages; VC++ programming language; explanation graph; interleaved plan; misleading actions; multiple plans; partial observability; partially ordered plan; plan recognition algorithm; Algorithm design and analysis; Artificial intelligence; Classification algorithms; Computer science; Educational institutions; Hidden Markov models; Probabilistic logic; ordering constraint; plan recognition; pruning;
Conference_Titel :
Computational Intelligence and Security (CIS), 2012 Eighth International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
978-1-4673-4725-9
DOI :
10.1109/CIS.2012.43