Menu
To provide intelligent solutions for computational/data intensive applications in the distributed environment.
To establish a world-class research centre to produce human resources in the area of Grid Computing.
This project is to propose a solution methodology for a missile defense problem involving the sequential allocation of defense resources over a series of engagements. The problem is computationally complex due to the presence of enormous state space. This paper proposes a reinforcement learning control approach for overcoming the complex state space issues. A new block architecture is proposed and implemented by using the LVQ-RBF multi agent Hybrid neural architecture. An artificial neural network (ANN) serves as the learning structure, and Q-learning as the learning method. The new architecture improvises the learning performance due to the local and global error criterion. The new architecture enables better simulation by increasing the number of assets and the number of categorization of the priorities used in simulation. The state space is explored by initial coarse partitioning and fine partitioning of the state space is performed by using the multi agent RBF neural network.
Apache Tomcat, JDK1.5, Protégé ontology editor, JSP, MATLAB, Oracle 9i, Clips