Dynamic decentralized task allocation algorithms for multi-agent systems using auctions and machine learning.
Master's research at The University of Texas at Austin in the research group of Efstathios Bakolas.
A paper resulting from these simulations is expected to be published by the end of 2020.
- For the dynamic task allocation, run
OptimalControl_DTA.m. - For the sensitivity analysis of the parameters, run
optimalControlParametersAnalysis.m.
Note: this repository is actively in progress and needs to be cleaned in the near future.
2D map of the dynamic task allocation (10 agents and 10 tasks) with associated reward, cost and utility:
