/TaxiProblem

The taxi problem solved with reinforcement learning (MAXQ value fonction decomposition)

Primary LanguagePython

TaxiProblem

  • Language : Python 3.6

  • Objective : solve the taxi problem with reinforcement learning (MAXQ value fonction decomposition).

  • For the moment, the taxi (green circle) has just to reach its target (the red small circle) as shown below. Only plain Q-Learning is implemented for the moment

  • How to use it ? Just run the main file
python3 main.py

You can change the variables in the file

variables.py

and the geometry of the domain by changing the variable

domain_file = "example_domain_3"

which is a path to the file example_domain_3 in the folder /domain.