Archive: Master's Student Projects (2006-2008)
Early explorations in AI during my Master's degree in Artificial Intelligence at the University of Groningen, The Netherlands. These course projects explored self-organizing systems, evolutionary algorithms, and cognitive modeling.
A Model for Ant Food Collecting Behavior in a Dynamical Environment
Course Project: Self-organizing Biological Systems (SOBS), University of Groningen, 2006
A. Goldhoorn, W. Lieffijn
PDF |
BibTeX
Extended NetLogo ant colony model implementing pheromone-based foraging in a Java simulation. Ants use pheromone trails to communicate food source locations, with emergent food distribution patterns arising from interaction between ant behavior and dynamic food growth. Demonstrates self-organization in a 10x larger environment (128×96 cells) with dynamic food patches.
Evolving RoboCode Tanks
Course Project: Robotica, University of Groningen, 2006
A. Goldhoorn
PDF |
BibTeX
Applied genetic algorithms to evolve behavior parameters for robot tanks in the RoboCode simulator. Implemented tournament selection, crossover, and mutation to optimize variables controlling robot behaviors like firing, movement, and target acquisition. Used behavior-based architecture combining subsumption and potential fields, with fitness evaluated through team battles. Demonstrates automatic optimization of multi-agent coordination strategies.
Mental Counting, Working Memory and Time
Course Project: Cognitive Modeling, University of Groningen
A. Goldhoorn
PDF |
BibTeX
ACT-R cognitive model of mental counting task requiring participants to maintain 2-3 counters in working memory while tracking toward a target value. Implements activation-based memory retrieval with decay, noise, and spreading activation. Models response timing using chunk activation and retrieval probability, capturing human performance patterns including errors and adaptation to task difficulty through interval time adjustments.