To build AD architecture, the equipment took as a model the way
human mental processes work. The developed architecture has two
levels: a Deliberative level and an Automatic one. In the Deliberative
level we find the skills that
require high computational time as a consequence of high level
reasoning. In the Automatic level lay the skills that interact with
robot sensors and actuators. In the AD architecture there are
several sequencers working in a concurrent way.
The navigation system was called AURON. Its autonomy is based on the interaction of four main components: the autonomy generation of an environment representation, the planning of a sequence of actions and perceptions which guide the robot from an initial event to a final event based on an optimization approach, the navigation that converts sequences in real movements and supervises all the process, and the last one, the relocalization that allows to place the robot again in the representation of the environment if it is lost. All these components have been implemented in the AD control architecture as deliberative skills.