Discrete event modeling and simulation: V-Lab/sup R/

Abstract

The need for modeling and simulation (M&S) is seen in many diverse applications such as multi-agent systems, robotics, control systems, software engineering, complex adaptive systems, and homeland security. With the emerging applications of multi-agent systems, there is always a need for simulation to verify the results before the actual implementation. Multi-agent simulation provides a test bed for several soft computing algorithms like fuzzy logic, neural networks (NN), probabilistic reasoning (stochastic learning automata, reinforcement learning), and evolutionary algorithms (genetic algorithms). Fusion of soft computing methodology with existing simulation tools yields several advantages in simulating multi-agent systems. Such a fusion provides a novel and systematic way of handling tune-dependent parameters in the simulation without altering the essential functionality and problem solving capabilities of soft computing elements. The fusion here is the extension of the capabilities of simulation tools with intelligent tools from soft computing. This paper proposes a methodology for combining the agent-based architecture, discrete event system and the soft-computing methods in the simulation of multi-agent systems and defines a framework called virtual laboratory (V-Lab/spl reg/) for realizing such multi-agent system simulations. Detailed experimental results obtained from simulation of robotics agents and wireless sensor network is also discussed.