ABSTRACT

This paper presents a biologically-inspired architecture for the control of autonomous underwater vehicles (AUVs) based on mathematical objects called integrons. The integron-based control architecture is illustrated by a case study of the hybrid control system design for an AUV whose mission is the installation of equipment in deepwater regions.

INTRODUCTION

The ocean environment imposes several restrictions on the design of underwater robots, due to unpredictable events and strong disturbances, that cannot be tackled using conventional control techniques. Therefore AUVs must possess a rich repertoire of properties and efficient mechanisms of reaction to accomplish their mission goals. Complex marine animals, like dolphins, whales and sharks, are multi-cellular biological organisms that exhibit autonomous behavior, including navigation, guidance, dynamic positioning and fault tolerance. This complex behavior emerges from biochemical networks, which are structures formed by the interactions of cells, through multiple signaling mechanisms, leading to organs, tissues and organisms. The Nobel Prize winner biologist Francois Jacob called these components, organized hierarchically according to nested structures, integrons (Jacob, 1970). The integron-based control architecture captures the essential characteristics of the biological networks, aiming at producing truly autonomous behavior for underwater vehicles.

THE INTEGRON-BASED ARCHITECTURE

Integrons are mathematical objects that emulate the autonomous behavior of multi-cellular biological organisms. They interact with each other through control restrictions and communication mechanisms. Each integron represents a component of a complex system and it owns an associated property of the system. The global behavior of such a system results not only from the individual behavior of the subsystems but also from the internal interaction among them and the external interaction with the environment. If the emergent observable behavior indicates some type of organization or a well defined set of functions, the interaction among the subsystems is of cooperative nature and it can be said that they are integrated (Ramos and Silva, 1999).

This content is only available via PDF.
You can access this article if you purchase or spend a download.