Reinforcement learning using potential field for role assignment in a multi-robot two-team game

Download

index.pdf

Date

2004

Author

Fidan, Özgül

Metadata

Show full item record

Item Usage Stats

229
views

96
downloads

In this work, reinforcement learning algorithms are studied with the help of potential field methods, using robosoccer simulators as test beds. Reinforcement Learning (RL) is a framework for general problem solving where an agent can learn through experience. The soccer game is selected as the problem domain a way of experimenting multi-agent team behaviors because of its popularity and complexity.

Subject Keywords

Computer engineering., Computer Hardware.

URI

http://etd.lib.metu.edu.tr/upload/12605724/index.pdf
https://hdl.handle.net/11511/14867

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Routing optimization methods for communication networks Demircan, Ahmet Emrah; Leblebicioğlu, Mehmet Kemal; Department of Electrical and Electronics Engineering (2005) This study discusses the routing optimization techniques and algorithms for communication networks. Preventing data loss on overloaded communication links and utilizing link bandwidths efficiently are the main problems of traffic engineering. Load balancing and routing problems are solved using both by heuristics such as genetic algorithms, and simulation techniques. These algorithms work on destination-based or flow-based routing techniques and mainly change the link weight system or try to select the best...
Vision based obstacle detection and avoidance using low level image features Senlet, Turgay; Halıcı, Uğur; Department of Electrical and Electronics Engineering (2006) This study proposes a new method for obstacle detection and avoidance using low-level MPEG-7 visual descriptors. The method includes training a neural network with a subset of MPEG-7 visual descriptors extracted from outdoor scenes. The trained neural network is then used to estimate the obstacle presence in real outdoor videos and to perform obstacle avoidance. In our proposed method, obstacle avoidance solely depends on the estimated obstacle presence data. In this study, backpropagation algorithm on mult...
An asynchronous system design and implementation of an FPGA Ayyıldız, Nizam; Güran, Hasan; Department of Electrical and Electronics Engineering (2006) Field Programmable Gate Arrays (FPGAs) are widely used in prototyping digital circuits. However commercial FPGAs are not very suitable for asynchronous design. Both the architecture of the FPGAs and the synthesis tools are mostly tailored to synchronous design. Therefore potential advantages of the asynchronous circuits could not be observed when they are implemented on commercial FPGAs. This is shown by designing an asynchronous arithmetic logic unit (ALU), implemented in the style of micropipelines, on th...
Video shot boundary detection by graph theoretic approaches Aşan, Emrah; Alatan, Abdullah Aydın; Department of Electrical and Electronics Engineering (2008) This thesis aims comparative analysis of the state of the art shot boundary detection algorithms. The major methods that have been used for shot boundary detection such as pixel intensity based, histogram-based, edge-based, and motion vectors based, are implemented and analyzed. A recent method which utilizes “graph partition model” together with the support vector machine classifier as a shot boundary detection algorithm is also implemented and analyzed. Moreover, a novel graph theoretic concept, “dominant...
Monte Carlo analysis of ridged waveguides with transformation media Ozgun, Ozlem; Kuzuoğlu, Mustafa (Wiley, 2013-07-01) A computational model is presented for Monte Carlo simulation of waveguides with ridges, by combining the principles of transformation electromagnetics and the finite methods (such as finite element or finite difference methods). The principle idea is to place a transformation medium around the ridge structure, so that a single and easy-to-generate mesh can be used for each realization of the Monte Carlo simulation. Hence, this approach leads to less computational resources. The technique is validated by me...

Citation Formats

Ö. Fidan, “Reinforcement learning using potential field for role assignment in a multi-robot two-team game,” M.S. - Master of Science, Middle East Technical University, 2004.