System Architecture, Development and Results of the Embry-Riddle Aeronautical University Maritime RobotX Platform

Abstract

Embry-Riddle Aeronautical University (ERAU) was selected as one of three schools to represent the United States in the inaugural Maritime RobotX Challenge, jointly sponsored by the Association for Unmanned Vehicle Systems International (AUVSI) and the Office of Naval Research (ONR). This international competition consisted of 15 teams from across the Pacific Rim; challenging them to develop a fully-autonomous surface vehicle using the 16foot high-performance Wave Adaptive Modular Vessel (WAM-V) Autonomous Surface Vehicle (ASV) platform to compete in Singapore, October 2014. The WAM-V ASV platform must accomplish multiple complex tasks fully autonomously, including buoy channel navigation, debris avoidance, plotting and updating of courses in real time, docking, and target identification. The novel system architecture utilized on the ERAU platform consists of multiple Ethernet enabled hardware components and software nodes running in parallel on separate computers to produce the complex behaviors required by the Maritime RobotX Challenge. These nodes include state estimation, health monitoring, object classification, map creation, mission planning, and trajectory planning. These systems and subsystems have been specifically designed and selected to maximize operational availability during component and software integration, which is of particular importance given the compressed one-year timeline of the project. To further this goal the hardware and software platform has been designed to allow for the upgrade and integration of both hardware and software in an iterative manner. This design philosophy, “complexity through iteration” maximizes test time and system capability. This paper discusses the development of the ERAU RobotX WAM-V ASV platform designated Minion, including an overview of the competition tasks, the modular hardware design, testing and the software architecture. The primary nodes covered here are obstacle identification, mapping, trajectory planning, and obstacle avoidance.