Event
M.S. Thesis Defense: Brian Beisel
Wednesday, May 21, 2014
3:00 p.m.
AVW 2328
Maria Hoo
301 405 3681
mch@umd.edu
ANNOUNCEMENT: M.S. Thesis Defense
Name: Brian Beisel
Committee:
Professor Gilmer Blankenship
ProfessorChristopher Davis
ProfessorNuno Martins
Date/Time: Wednesday, May 21, 2014 at 3pm-5pm
Location: AVW 2328
Title: Distributed Maze Solving by Cooperative Robotic Platforms
Abstract:
Problem solving based on sensor created area maps is a challenging problem that can benefit from a multi-robot approach. Cooperative problem is most eloquently designed through distributed services and systems. This thesis designs and implements a full distributed maze solving solution using simulated robotic sensor platforms. A distributed spatial communication system was developed and tested as a contributing element of the maze solving solution. Autonomous algorithms for communication, cooperation, and navigation were constructed and tested through simulation in maze solving tests. Working with an assumed map creating technology in tandem with the aforementioned developed technologies resulted in an effective complete solution. It was found through simulation and mathematical analysis that multiple cooperative robot platforms can result in significant performance improvements. A great deal of future work is recommended to address imperfect mapping complications.
Problem solving based on sensor created area maps is a challenging problem that can benefit from a multi-robot approach. Cooperative problem is most eloquently designed through distributed services and systems. This thesis designs and implements a full distributed maze solving solution using simulated robotic sensor platforms. A distributed spatial communication system was developed and tested as a contributing element of the maze solving solution. Autonomous algorithms for communication, cooperation, and navigation were constructed and tested through simulation in maze solving tests. Working with an assumed map creating technology in tandem with the aforementioned developed technologies resulted in an effective complete solution. It was found through simulation and mathematical analysis that multiple cooperative robot platforms can result in significant performance improvements. A great deal of future work is recommended to address imperfect mapping complications.
