Prev-Talk | Next-Talk | All-Talks | Talks-Sorted

Talk-06: Autonomy and Collaboration Research using MOOS with the Yellowfin UUV at the Georgia Tech Research Institute

Andrew Melim, Michael Novitzky, Paul Varnell, Kevin DeMarco, Tom Collins, Mick West, Georgia Tech Research Institute

Georgia Tech Research Institute's Yellowfin is a low-cost, lightweight AUV designed for use in mutli-vehicle collaborative missions. It utilizes the MOOS middleware in addition to IvP-Helm to enable rapid development of the platform as well as autonomous behavior research. MOOS plays a large component in continuing research. IvP-Helm is a key component in developing methods of recognizing behaviors in multi-vehicle missions should there be a loss of communications or vehicle malfunction. Identifying and compensating for behavior changes due to such events will lead to more robust collaboration in multi-AUV missions. Additionally, the research also focuses on passivity based controls over acoustic communications with an unknown discrete time-varying delay in order to improve the stability of vehicle control. Finally, underwater navigation through the use of a high frequency imaging sonar underwater SLAM is also currently being researched on the Yellowfin platform.

Use of the MOOS-IvP package in addition to pAcommsHandler provides a large backbone of the vehicle's software programming. This presentation will briefly overview our research and discuss how MOOS influences both the hardware and software development of the Yellowfin AUV. Roadblocks encountered using MOOS in the process of our research will be discussed as well as suggestions for improvements.

Categories:

  • The Yellowfin UUV
  • MOOS-IvP
  • Unmanned Underwater Vehicles (UUVs) / Autonomous Underwater Vehicles (AUVs)
  • Underwater SLAM
  • Acoustic Communications