Transputer-Based Distributed Processing For Underwater Robotic Vehicle Control

Abstract

High-precision, closed-loop control of underwater robotic vehicles presents unique computing requirements. This paper outlines a vehicle computer design philosophy based on transputers, microprocessors specially designed for parallel processing. Distributed processing can be seen as a subset of parallel processing, and transputers can be adapted to serve as high-performance distributed controllers for underwater vehicles, both tethered and autonomous. Two transputer-based designs are presented, Hylas, a tethered research vehicle, and ABE, an autonomous survey system. Data is presented to show the low-latency and high update-rate of vehicle controllers can be achieved with transputers as distributed controllers.