Talk-01: A Flexible, Low-Cost MOOS-IvP Based Platform for Marine Autonomy Research
Damian Manda, NOAA Office of Coast Survey, NOAA-UNH Joint Hydrographic Center, Ocean Engineering Program, University of New Hampshire
Andrew D’Amore, Systems Engineering Program, University of New Hampshire
May-Win Thein, Ocean Engineering Program, University of New Hampshire
Andrew Armstrong, NOAA Office of Coast Survey, NOAA-UNH Joint Hydrographic Center
As part of an effort to research collaborative autonomy behaviors between multiple unmanned surface and underwater vehicles, a hardware and software system has been developed for control of autonomous surface vessels (ASVs). This system is designed to be flexible in application to diverse platforms and ability to execute complex missions. In order to facilitate duplication across many deployments, the cost of the full system is minimized by leveraging mass produced, open source hardware and software.
MOOS-IvP is used as the central data assimilation and decision making software. Stock supplied and newly developed IvP behaviors are used to plan the trajectory for the ASV and can be customized to suit platform requirements. A graphical interface is available for setting missing configuration parameters to simplify deployment by those not fully versed in MOOS mission file creation. Sensor data is assimilated through either MOOS interface drivers or using ROS software and then passed to MOOS for use by the IvP helm and navigation controller. Incorporating ROS allows flexibility in sensor selection as many drivers already exist in the community and can be quickly adapted to this autonomy system.
The hardware for the system is designed to be housed in a single watertight box. Onboard processing has been deployed on the low cost BeagleBone Black and Raspberry Pi 2 platforms. Low level sensor input and control output as well as fail-safes and human remote control are handled by an independent Arduino microcontroller. Position and orientation input can be accepted from an existing source on the vessel or use a MEMS INS for simple deployments. The autonomy system has been implemented on multiple small vessels including those with both gas and electric engines. DOWNLOAD
- Payload Autonomy Platform/Interface
- Cross Domain UxVs
- Command and Control, Mission Planning