A Fuzzy Path Planning System Based on a Collaborative Reinforcement Learning

摘要

In this paper, we present a novel collaborative Q-Learning based path planning system using Holonic Multi Agent System architecture, and the Fuzzy Inference System, to be used in autonomous mobile robots, represented as head-holons, in order to find the optimal path among any starting point, and a goal in a to 2D grid environment. The decision of the navigation is provided by a fuzzy system controller and then it is verified by a sensor validator.