Hardware and software architecture for execution control of an autonomous mobile robot

The concepts of a general hardware and software architecture for the execution control of autonomous mobile robots are discussed. These concepts were applied to build HILARE II, a new experimental mobile robot. By definition, a control structure is composed of the software elements responsible for the execution control of robot actions. In this approach, the control structure of a complex robotic system can be viewed as successive software layers, built on the top of the physical robot. The first layer (logical robot) implements the access to the physical robot by higher level software layers. The next layer (module substratum) consists of a group of functional modules which cooperate to accomplish the tasks required by a central controller. Each module is capable of executing specific services, by acting either on the logical robot or on other modules. The module substratum represents the distributed control component of the system. A central controller verifies the robot's and the environment's state evolution, ensures the applicability of planned actions, starts the execution of these actions, and defects/recovers from possible failures.<
>