ACROSS CoE - Centre of Excellence for Autonomous and Cooperative Robotic Systems, IEEE Hrvatska sekcija, Odjel za robotiku i automatizaciju, ZCI-DATACROSS i H2020 projekt "EXCELLABUST - Excelling LABUST in marine robotics" vas pozivaju na predavanje
"Underwater Robot Navigation and SMaRC"
koje će održati Prof. John Folkesson, KTH Royal Institute of Technology, Stockholm, Švedska. Predavanje će se održati u utorak, 17. travnja 2018. godine, u 11.00 sati, u zavodskom seminaru ZARI-ja, zgrada C, 9. kat.
Više o predavaču i predavanju možete pročitati u opširnijem sadržaju obavijesti.
Navigating underwater presents challenges well beyond those seen in indoor and outdoor robotics. There is poor visibility, sparse landmarks, strong disturbances, no gps and no communications.
Leveraging progress in SLAM to solve the underwater navigation problem has had only limited success due to the vastness of the oceans and the sparseness of distinguishing features. The talk will present some earlier work done on underwater SLAM and localization using forward looking sonar for mine counter measures. The Swedish Maritime Robotics Centre, SMaRC, is also very focused on the problem of navigation. The centre will be briefly introduced and the challenging scenarios being addressed will be presented. The problem of navigating under an Antarctic ice shelf in particular is being targeted by SMaRC. This is needed to collect dense and complete data on the energy flows into the under ice cavity from relatively warm dense ocean currents.
Understanding these energy flows is critical to prediction of ice melt and ocean level rise. Navigation for a week or more at a time in an unknown under ice environment will push the boundaries of the possible. Properly geo-referencing the data collected and accurately mapping the under ice topography will provide missing pieces to the puzzle of heat transfer between the water and ice.
John Folkesson is an Associate Professor at KTH - Royal Institute of Technology where he does research on robotics. After obtaining his robotics PhD in 2006 from KTH, he was a researcher Massachusetts Institute of Technology. At MIT he worked on autonomous underwater vehicles until, 2010 when he returned to KTH. His research has focused on interpretation of robot sensor data, including simultaneous localization and mapping, SLAM.