Skip to content

Reinforcement Learning + traffic microsimulation (via SUMO). Uses Ray RLLIB and forces SUMO into the OpenAI Gym Framework

Notifications You must be signed in to change notification settings

mschrader15/reinforcement-learning-sumo

Repository files navigation

Reinforcement Learning + SUMO

Applying reinforcement learning to traffic microsimulation (SUMO)

A minimal example is available in the example folder

Very much a WIP

Structure

This project follows the structure of FLOW closely. I only chose to diverge from FLOW because it abstracted the XML creation for SUMO. For me, this repository plugs in to a greater code-base, that turns real-world ITS data into SUMO traffic demand and traffic light operation. Those tools work ultimately by writing to SUMO XML files, and we didn't want to convert to the FLOW pythonic framework.

Results

Full breakdown: https://maxschrader.io/reinforcement-learning-and-sumo

video on Youtube

ES vs. PPO

RL-Controlled Traffic Signals vs. Calibrated Real-World Simulation

Real World Traffic Signals during Simulation Period

RL-Controlled Traffic Signals during Simulation Period

About

Reinforcement Learning + traffic microsimulation (via SUMO). Uses Ray RLLIB and forces SUMO into the OpenAI Gym Framework

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published