SELF-ADAPTING REACTIVE AUTONOMOUS AGENTS
M. Andrecut, Mazhar Ali · International Journal of Modern Physics B · 2000
This paper describes a new self-adapting control algorithm for reactive autonomous agents. The architecture of the autonomous agents integrates the reactive behavior with reinforcement learning. We show how these components perform on-line adaptation of the autonomous agents to various complex navigation situations by constructing an internal model of the environment. Also, a discussion on cooperation and coordination of teams of agents is presented.