RL4OCC is a project that utilizes Deep Reinforcement Learning (DQN) to solve assembly sequence problems. The project employs both Genetic Algorithm (GA) and DQN to optimize the assembly sequence, reduce the number of collisions during assembly, and improve assembly efficiency.
Below is an overview of the project's directory structure and important files:
rl4occ/
├── assembly.py
├── attention_q_net.py
├── data
├── data_process.py
├── dqnlearn.py
├── env.py
├── LICENSE
├── main.py
├── misc
├── misc.py
├── model
├── optimise.py
├── README.md
├── reference
├── replay_buffer.py
└── utils.py
Run the following command in the project root directory to install the required Python libraries:
pip install -r requirements.txt
Running DQN Training Execute the following command to start DQN training:
python main.py
Running Genetic Algorithm Optimization Use the following command to run the Genetic Algorithm optimization process:
python optimise.py
This project is licensed under the MIT License.