Faculty of Informatics Vienna University of Technology Institute of Computer Aided Automation PRIP Home PRIP Home
Personal tools
You are here: Home Teaching PRIP-Talks David Major: Observation of the Dynamics of Autonomous Underwater Vehicles
Navigation
 

David Major: Observation of the Dynamics of Autonomous Underwater Vehicles

Abschlussvortrag Informatikpraktikum

What
  • Presentation
When May 12, 2009
from 04:15 pm to 04:35 pm
Where Sem 183/2
Add event to calendar vCal
iCal
Within this research project I propose a Cluster analysis approach based on statistical tests which is applied on data from Autonomous Underwater Vehicles. Autonomous Underwater Vehicles are robotic devices, which are used for observation of the ocean floor. These devices belong to the efficient and productive technologies which enable the reporting on the interior of the ocean. Within the scope of a previously developed project by the CIS Department of the University of Strathclyde the stochastic learning algorithms, Kohonen Self Organizing Map and Expectation Maximization, will be used to learn a discrete observation model (Hidden Markov Model) of robotic devices like Autonomous Underwater Vehicles. The goal is to enhance the learning of the discrete observation model by using Cluster analysis techniques for Cluster reorganization and so making it possible to build higher level representations such as a Partially Observable Markov Decision Process. For that purpose I applied statistical tests, such as MANOVA, Two-Sample Hotelling?s T-Square test and a Confidence Interval test as a Cluster analysis step on the results of the stochastic learning algorithms. A reduction of the number of Clusters could be achieved by merging similar ones. With the Two-Sample Hotelling?s T-Square statistical test I could reach the highest Cluster reduction with significance levels of 0.1%, 1%, 5% and 10% and so prepared the building of high level representations on datasets with high number of Clusters (around 500 Clusters) as a result.
Document Actions