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.

  

RoboCup: Dynamic Role Assignment within the B-Human Framework

Author(s):

K. Steich
Conference/Journal:

Semester Thesis, HS14 (10382)
Abstract:

There is currently an increased interest in systems of physical autonomous robots exhibiting decentralized cooperative behavior operating in noisy and dynamic environments, using limited communication. Robotic soccer teams in the Standard Platform League of the RoboCup can be considered as such systems. This work focuses on developing dynamic role assignment for such a team, namely the Standard Platform League robotic soccer team of the ETHZ, the Z-Knipsers. Drawing on the work of other robotic soccer teams, we develop a dynamic role assignment approach that uses utility functions together with a formation system and can be formulated as an iterated form of the well-studied optimal assignment problem. Robots will be iteratively assigned to distinct roles, or more specifically positions in a formation based on a utility function that determines how well suited they are for each role using a greedy algorithm. Formations are designed based on di↵erent strategies and new roles are devised. Details about the implementation of this dynamic role assignment approach within the B-Human code framework are given. Simulation results and some experiments with real robots show that the implementation functions as intended and the approach displays a promising performance compared to static role assignment.

Supervisors: Khoa Nguyen, John Lygeros

Year:

2015
Type of Publication:

(13)Semester/Bachelor Thesis
Supervisor:

K. Nguyen

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2015:IFA_5123
}
Permanent link