Incremental Evolution of Autonomous Robots for a Complex Task

An incremental approach is used to develop the control system of an autonomous robot to approach toward the target object by avoiding obstacles in an environment. The approach consists of two-stage evolution. In the first-stage, controllers are evolved to avoid obstacles in an environment. The final population of the first-stage evolution is then used as the initial population for the second-stage evolution. Controllers are evolved in the second-stage to approach toward the target object by avoiding obstacles in the environment. We compare the performance of the incremental approach with that of conventional approach, an onestage evolutionary approach. It is found that the performance of incremental approach is better than that of conventional approach.