Glider Programming Architecture

Collaborators

Publications

• Towards a Resource-Aware Programming Architecture for Smart Autonomous Underwater Vehicles , H.C. Woithe, D. Tilkidjieva, and U. Kremer, Technical Report, Department of Computer Science, Rutgers University, DCS-TR-637, June 2008.
  
  
• A Programming Architecture for Smart Autonomous Underwater Vehicles , H.C. Woithe and U. Kremer, IEEE International Conference on Intelligent Robots and Systems (IROS'09), St. Louis, Missouri, October 2009.
  
  
• Slocum Glider Energy Measurement and Simulation Infrastructure, H.C. Woithe, I. Chigirev, D. Aragon, M. Iqbal, Y. Shames, S. Glenn, O. Schofield, I. Seskar, and U. Kremer, IEEE Oceans Conference, Sydney, Australia, May 2010.
  
  
• An Interactive Slocum Glider Flight Simulator, H.C. Woithe and U. Kremer, IEEE Oceans Conference, Seattle, WA, September 2010.
  
  
• Using Slocum Gliders for Coordinated Spatial Sampling, H.C. Woithe and U. Kremer, IEEE Oceans Conference, Santander, Spain, June 2011.
  
  
• Improving Slocum Glider Dead Reckoning Using a Doppler Velocity Log, H.C. Woithe, D. Boehm, and U. Kremer, IEEE Oceans Conference, Kona, Hawaii, September 2011.
  
  
• A Lightweight Scripting Engine for the Slocum Glider, H.C. Woithe and U. Kremer, IEEE Oceans Conference, Kona, Hawaii, September 2011.
  
  
• Opportunities to Parallelize Path Plannning Algorithms for Autonomous Underwater Vehicles, M. Eichhorn and U. Kremer, IEEE Oceans Conference, Kona, Hawaii, September 2011.
  
  
• Parallelization of Path Planning Algorithms for AUVs - Concepts, Opportunities, and Program-Technical Implementation, M. Eichhorn, H.C. Woithe, and U. Kremer, IEEE Oceans Conference, Yeosu, South Korea, May 2012.
  
  
• Enabling Computation Intensive Applications in Battery-Operated Cyber-Physical Systems, H.C. Woithe, W. Brozas, C. Wills, B. Pichai, U. Kremer, M. Eichhorn, and M. Riepen, MARC Symposium, Toulouse, France, July 2012.
  
  
• Assessing Automated and Human Path Planning for the Slocum Glider, H.C. Woithe, M. Eichhorn, O. Schofield, and U.Kremer, 18-th International Symposium on Unmanned Untethered Submersible Techology (UUST 2013), Portsmouth, New Hampshire, August 2013.
  
  
• Comparison of Guidance Modes for the AUV "Slocum Glider" in Time-Varying Ocean Flow, M. Eichhorn, H.C. Woithe, and U.Kremer, IEEE Oceans Conference, Taipei, Taiwan, April 2014.
  
  
• Design and Implementation of an Energy Aware Programming Framework for Autonomous Underwater Vehicles, H.C. Woithe, Ph.D. Thesis, Rutgers University, May 2014.
  
  
• Feature Based Adaptive Energy Management of Sensors on Autonomous Underwater Vehicles, H.C. Woithe and U. Kremer, Ocean Engineering, Volume 97, March 2015, pages 21-29.
  
  
• Developing Coordinated Communities of Autonomous Gliders for Sampling Coastal Ecosystems , O. Schofield, C. Jones, J. Kohut, U. Kremer, T. Miles, S. Travis, G. Saba, D. Webb, and S. Glenn, Marine Technology Society Journal, Volume 49, Number 3, May/June 2015, pages 9-16.
  
  
• TrilobiteG: A Programming Architecture for Autonomous Underwater Vehicles, H.C. Woithe and U. Kremer, ACM SIGPLAN/SIGBED Conference on Languages, Compilers, Tools and Theory of Embedded Systems (LCTES'15), Portland, Oregon, June 2015.
  The TrilobiteG simulation environment is available at algaesim.cs.rutgers.edu .

News

Missions

• September 27, 2012 -- MRI-1: Eighth mission, one deployment. The deployment was in approx. 200 feet (65 meters) of water off the coast of Belmar, New Jersey. Our onboard computer, AvBot, took control of the glider and used adaptive sampling to trigger the two on-board bio-puck sensors only within or close to the thermocline. During the first diving pass, the thermocline is detected. Subsequently, the bio-pucks (red dots) are activated only close to the thermocline. The thermocline temperature reading are shown in blue. The experiment was repeated several times with similar results. The used triggering algorithm has been refined over the version that we used in the July 17 deployment (see below).
  
  
  
• July 17, 2012 -- MRI-1: Seventh mission, one deployment. The deployment was in approx. 115 feet (35 meters) of water off the coast of Belmar, New Jersey. Our onboard computer, AvBot, took control of the glider and used adaptive sampling to trigger the two on-board bio-puck sensors only within or close to the thermocline. During the first diving pass, the thermocline is detected. Subsequently, the bio-pucks (red dots) are activated only close to the thermocline. The thermocline temperature reading are shown in blue. The experiment was repeated twice with similar results.
  
  
  
  
  
  
  
• October 18, 2011 -- MRI-2: Sixth mission, one deployment. This first deployment of MRI-2 was 5 miles off the coast of Atlantic City in approximately 55 feet of water. While submerged, a "surfacing" command was sent to the glider via an acoustic modem from our deployment vessel. MRI-2's WHOI acoustic modem received the command, passed it on to our onboard computer, which then issued an instruction to the flight controller to resurface. We sucessfully repeated this experiment.
  
  
  
  
  
  
  
• December 9, 2010 -- Outreach Program: Sixth Grade Science Class at Maplewood Middle School, New Jersey.
  As part of a lesson unit covering density in sixth grade science, we used a 8-foot wave tank to illustrate how water of different densities mix, such as fresh and salt water. The experiments show how water of different salinities and therefore different densities form distinct layers. Food colors were used to allow these layers to be observed.
  
  We also illustrated how water density is used by Autonomous Underwater Vehicles such as the Slocum glider. The Slocum glider is a buoyancy driven vehicles that changes its overall density to propel itself through the water column. Students were able to inspect an actual Slocum glider, and use our interactive glider simulator to plan and fly missions across the continental shelf off the coast of New Jersey. Overall, we covered five periods of science and were able to reach over 80 students. We would like to thank Mark Terenzi, a sixth grade science teacher of Team A at Maplewood Middle School, for all his help and support.
  
  
  Photos by Hans Christian (Chris) Woithe
  
  
  
• August 3 through September 1, 2010 -- RU01: Fifth mission, one deployment. RU01 is equipped with an updated, more power-efficient version of our new power measurement infrastructure. The glider went out to the edge of the continental shelf twice, collecting power measurements of the buoyancy pump and other motors. In addition to obtaining data, the experiment showed that our infrastructure is efficient enough to be deployed in a 4 week mission.
  
  
  
  Flight path of RU01 over a period of 4 weeks.
  
  
  .
  
  
• February 3 through 9, 2010 -- RU07: Fourth mission, one deployment. RU07 was equipped with our new power measurement infrastructure. This was the first time that we flew without a safety net, i.e., without a buoy attached to the glider. The goal was to reach the edge of the continental shelf, take power measurements at depths of up to 100 meters, and verify the overall functionality of the board under field conditions. Unfortunately, the weather did not cooperate. A Nor'easter forced the glider northward as can be seen in the mission flight map below. Overall, the mission was still a success since we got power readings up to a depth of 32 meters and new insights into our current infrastructure.
  
  
  
  This movie spans a 6 day time period, starting from the RU07 deployment on February 3, to her pick-up on February 9.
  
  
  
• September 21, 2009 -- RU10: Third mission, one deployment. RU10 was equipped with a new power measurement infrastructure that is able to measure the in-flight energy consumption of different sensors and motors. The goal of this mission was to calibrate our new power measurement board. The board was developed in collaboration with Ivan Seskar , a researcher at Winlab , and his students Yuriy Shames, Ilya Chigirev, and Murium Iqbal. The picture below shows RU10 on the surface before starting its one hour mission. The glider is connected to a buoy in case of system failure.
  
  
  
  
  
  
• September 4, 2008 -- RU15: Second mission, one deployment. Successful tracking of a thermocline at around 15 meter depth off the coast of New Jersey as shown in Figure labeled Seabird CTD . The flight path of the glider is shown in Figure labeled Glider Profile . After 1500 seconds, the old programming architecture gained control again, and the glider surfaced.
  
  
• July 31, 2008 -- RU20July31.2008.mov (61MB): First mission off the coast of New Jersey with two glider deployments. The video shows the second deployment of the Slocum glider RU20. RU20 was equipped with our new programming architecture.
  
  

Research Group Wiki (limited access)