Integrating expert system CLIPS into DUNE for AUV control

In this paper, we motivate the need for a flexible and robust system architecture for autonomous underwater vehicles. We present a novel concept for integrating the expert system CLIPS into an AUV navigation environment DUNE. Our approach provides complete tool support for automatic source code generation and therefore allows fast and target-oriented development of control software for AUVs. We present the design of the architecture comprised of DUNE and CLIPS. In an application example we describe each step of the software development with implementation details. We show the benefits of such an expert system for a flexible AUV control. For evaluation we measure the execution time of the system and the number of fired rules of our expert system. By that we demonstrate that the approach is suited for embedded hardware of an AUV.