Balancing exploration with exploitation-solving mazes with real numbered search spaces

A hybrid architecture, called EXP1, automatically balances exploration and exploitation to solve mazes with large real numbered search spaces. It employs a genetic algorithm (GA) to search and optimise each movement. The GA fitness function is supplied by a radial basis function (RBF) neural network which acts as an adaptive heuristic critic (AHC). Over successive trials it learns the V-function, a continuous mapping between real numbered positions in the maze and the value of being at those positions. EXP1 solved all the mazes with which we tested it and proved to be quite robust to changes in internal parameters