The software architecture of the mu-Net infrastructure aims to provide an integrated software platform for sensing, communication and robotics. The objectives include: (a) Execute mobile missions including sensing, localization, real-time remote command & control, etc.; (b) Support communications of different types of data traffic: asynchronous data traffic (e.g., sensing data, navigation data, environment data, network and system monitoring, probing data), and time-bounded data (mission command and control, emergency event reports, on-line reconfiguration); (c) Be extensible and flexible to different hardware capacities or software functions; (d) Support simulation, emulation, and real field experiments; and (e) Support optimizations under various resource constraints.
The design of the software is illustrated below. In this architecture, we integrate underwater networking capacity into ROS (Robot Operating System)  and MOOS (Mission Oriented Operating Suite) . ROS is an open source robotics middleware suite with a collection of software for robot software development. It is a well know software suite in the robotics domain with a large user space and numerous resources. MOOS, on the other hand, is popular in the marine robotics community. It is a C++ cross platform that has been used widely for various autonomous underwater vehicles. Leveraging the resources of ROS and MOOS, the mu-Net software can have access to a plethora of application software, from sensing, localization, command & control, to navigation. In addition, the communication and networking capacity is provided by Aqua-Net. It is a highly flexible, reconfigurable and customizable protocol stack, designed specifically for underwater networks.
ROS + Aqua-Net Demo
MOOS + Aqua-Net Demo
- Morgan Quigley, Brian Gerkey, Ken Conley, Josh Faust, Tully Foote, Jeremy Leibs, Eric Berger, RobWheeler, and Andrew Ng, ROS: An Open-source Robot Operating System, in Proceedings of ICRA Workshop on Open Source Software, 2009
- Paul Newman, An overview of the MOOS Architecture, MOOS project website.
- Zheng Peng, Zhong Zhou, Jun-Hong Cui and Zhijie Shi,