propose a novel RL method for multi-intersection traffic signal control on the arterials.
DQN，每个agent控制自己的Intersection；互相之间无信息传递
- draw a connection on the design between reinforcement learning with conventional transportation control methods.
- the first time the individual RL model automatically achieves coordination along arterial without any prior knowledge
训练过程运行runexp.py，对训练好的模型进行replay.txt进行生成使用replay.py。
Before running above codes, you may need to install following packages or environments:
- Python 3.6
- Keras 2.2.0
- Tensorflow 1.13.0