This repo contains code accompaning the paper: Zhi Wang, Chunlin Chen, Han-Xiong Li, Daoyi Dong, and Tzyh-Jong Tarn, "Incremental reinforcement learning with prioritized sweeping for dynamic environments", IEEE/ASME Transactions on Mechatronics, 2019. It contains code for running the incremental learning tasks with a discrete state-action space, including the simple maze and complex maze domains, as stated in the paper.

This code requires the following:

• python 3.*
• gym

• For the simple maze domain, data is generated from myrllib/envs/simple_maze.py
• For the complex maze domain, data is generated from myrllib/envs/complex_maze.py

• For example, to run the code in the simple maze domain with epsilon-greedy strategy, just run the bash script ./simple_maze_epsilon.sh, also see the usage instructions in the script and main.py
• When getting the results in the folder output/*, plot the results using data_process.py. For example, the results for ./simple_maze_epsilon.sh is as follow:

Also, the results for other bash scripts are shown in exp/*

To ask questions or report issues, please open an issue on the issues tracker, or email to [email protected].