Design of Task Based Control Architecture for Personal Robots
vi, 35 p.
- 원문 URL
As robot's application fields increase with much more complicated environments and tasks, robots need more appropriate control architectures. Previous control architectures have fixed configurations which are optimized for specific applications or with limited flexibility, so that they fail to provide enough flexibility for various robot kinematic configurations as well as various tasks. To this end, we propose a new control architecture called as "Supervised Hybrid Architecture (SHA)". SHA is based on supervised organization and distributed arbitration of hybrid controls of reconfigurable deliberative and reactive modules. SHA is composed of upper level hybrid control for high-level intelligence to interact with human and to plan task, as well as lower level hybrid control to allow low-level intelligence of prompt reaction in each robot configuration module. By this double layers of hybrid controller, we could solve complexity from complex robot configurations as well as provide enough flexibility necessary for so many different tasks. In addition, it is very easy to add or remove robot configuration modules. The proposed architecture are implemented and tested for guide tasks to show how it works successfully.