An Incremental Approach in Evolving Robot Behavior

摘要

This paper describes an incremental evolutionary approach used in the development of a suitable neural controller for achieving robust obstacle avoidance behavior, which is then further fine-tuned towards a wall following one for a simple mobile robot. The incremental approach mainly involves an alteration of the environment in which the evolution takes place as well the fitness function used in the genetic algorithm. This approach has been seen to be more fruitful than a single direct approach. Interesting behaviors have evolved from this incremental approach.