Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Autonomous surveillance with a Pan-Tilt-Zoom Camera

Author(s):

Samuel Aufdenblatten
Conference/Journal:

Semester Thesis HS 11 (10118)
Abstract:

Staring from a preliminary developed Matlab tracking algorithm, a real time surveillance framework for a "`Ullisse"' pin hole camera has been developed in C/C++. As object to track a Khepera mobile robot is used. In a rst step, the position and the orientation of the robot is calculated by processing the images taken with the camera. Then a probability distribution of the robot's position is estimated. Based on this distribution, a stochastic reach-avoid problem is posed, which is solved by a dynamic programming algorithm. To be able to multiply the occurring matrices (size around 50'000x2500) in few milliseconds, the high performance Intel Math library (MKL) is used. The solution to the problem is a strategy to steer the camera over the horizon, so that the probability of ful lling an objective is maximized. It was possible to successfully test the framework on a real test bed, where a Khepera mobile robot was driving around with a velocity of 0:3m s .

Year:

2011
Type of Publication:

(13)Semester/Bachelor Thesis
Supervisor:

N. Kariotoglou

No Files for download available.
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2011:IFA_4128
}
Permanent link