Genetic algorithms (GAs) are stochastic, parallel search algorithms based on the mechanics of natural selection, the process of evolution [4, 3]. GAs were designed to efficiently search large, non-linear, poorly-understood search spaces where expert knowledge is scarce or difficult to encode and where traditional optimization techniques fail. They are flexible and robust, exhibiting the adaptiveness of biological systems. As such, GAs appear well-suited for searching the large, poorly-understood spaces that arise in design problems; specifically designing control strategies for mobile robots.
CHC, the non-traditional genetic algorithm used in this paper, differs from traditional GAs in a number of ways [2]:
Case-based reasoning (CBR) is based on the idea that reasoning and explanation can best be done with reference to prior experience, stored in memory as cases [12]. When confronted with a problem to solve, a case-based reasoner extracts the most similar case in memory and uses information from the retrieved case and any available domain information to tackle the current problem. This paper uses the basic tenet of CBR -- the idea of organizing information based on ``similarity'' -- to help augment genetic algorithm search.