Architecture of the Texas A&M Autonomous Underwater Vehicle Controller

Abstract

Presents the software and hardware architectures of the autonomous underwater vehicle controller (AUVC) developed at Texas A&M University. It is a controller for a long range, highly reliable UUV. Capabilities include mission planning/replanning, path planning, energy management, collision avoidance, threat detection and evasion, failure diagnosis and recovery, radio communication, navigation, and recovery from its internal faults. In its first version, functions were partitioned among eighteen loosely coupled processes. Rule-based systems performed mission management and fault diagnosis, while algorithmic control systems were used for lower-level control. The original AUVC software was designed for a network of sixteen processors in planar-2 configuration, with redundant communication paths. A software component provided reliable distributed computing. The controller was tested using a simulated generic vehicle that contained twenty-one subsystems. Tests on the Large Diameter UUV (LDUUV), usin a six-processor version of the AUVC, are reported.