An Artificial Brain System of Maze-like Robot
Xinyuan Li · Control Engineering of China · 2010
An artificial brain system of a maze-like robot is presented,which comprises a perception unit and a decision-making unit.The perception module is based on ART1 neural network and trained to identify the signposts of the maze.The decision-making units is based on behavior probability matrix,and uses reinforcement learning to update the action strategy.The maze has signposts at each intersection,which are 2-D symbols with noise.In the simulation tests,the robot moves randomly in the maze.By adjusting the parameters of the experiment,the robot will eventually pass through the maze after a learning process during the self-exploration.The simulation result shows that the artificial brain system can be self-organized to make sense of the signposts and successfully guide the robot through the maze.It is also has a positive impact on patrol robot and rescue robots based on signposts navigation.