https://github.com/cityflow-project/CityFlow

A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
https://github.com/cityflow-project/CityFlow

multiagent-reinforcement-learning multiagent-systems traffic-signal-control traffic-simulation

Last synced: 3 months ago
JSON representation

A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario

• Host: GitHub
• URL: https://github.com/cityflow-project/CityFlow
• Owner: cityflow-project
• License: apache-2.0
• Created: 2019-02-26T04:16:52.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2022-02-17T14:45:39.000Z (over 2 years ago)
• Last Synced: 2024-04-14T04:58:15.836Z (7 months ago)
• Topics: multiagent-reinforcement-learning, multiagent-systems, traffic-signal-control, traffic-simulation
• Language: C++
• Homepage: https://cityflow-project.github.io
• Size: 332 KB
• Stars: 735
• Watchers: 19
• Forks: 168
• Open Issues: 39
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE.txt

Awesome Lists containing this project

• open-sustainable-technology - CityFlow - A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario. (Consumption / Mobility and Transportation)

README

        
CityFlow

============

.. image:: https://readthedocs.org/projects/cityflow/badge/?version=latest

    :target: https://cityflow.readthedocs.io/en/latest/?badge=latest

    :alt: Documentation Status

.. image:: https://dev.azure.com/CityFlow/CityFlow/_apis/build/status/cityflow-project.CityFlow?branchName=master

    :target: https://dev.azure.com/CityFlow/CityFlow/_build/latest?definitionId=2&branchName=master

    :alt: Build Status

CityFlow is a multi-agent reinforcement learning environment for large-scale city traffic scenario.

Checkout these features!

- A microscopic traffic simulator which simulates the behavior of each vehicle, providing highest level detail of traffic evolution.

- Supports flexible definitions for road network and traffic flow

- Provides friendly python interface for reinforcement learning

- **Fast!** Elaborately designed data structure and simulation algorithm with multithreading. Capable of simulating city-wide traffic. See the performance comparison with SUMO [#sumo]_.

.. figure:: https://user-images.githubusercontent.com/44251346/54403537-5ce16b00-470b-11e9-928d-76c8ba0ab463.png

    :align: center

    :alt: performance compared with SUMO

    Performance comparison between CityFlow with different number of threads (1, 2, 4, 8) and SUMO. From small 1x1 grid roadnet to city-level 30x30 roadnet. Even faster when you need to interact with the simulator through python API.

Screencast

----------

.. figure:: https://user-images.githubusercontent.com/44251346/62375390-c9e98600-b570-11e9-8808-e13dbe776f1e.gif

    :align: center

    :alt: demo

Featured Research and Projects Using CityFlow

---------------------------------------------

- `PressLight: Learning Max Pressure Control to Coordinate Traffic Signals in Arterial Network (KDD 2019) `_

- `CoLight: Learning Network-level Cooperation for Traffic Signal Control `_

- `Traffic Signal Control Benchmark `_

- `TSCC2050: A Traffic Signal Control Game by Tianrang Intelligence (in Chinese) `_ [#tianrang]_

Links

-----

- `WWW 2019 Demo Paper `_

- `Home Page `_

- `Documentation and Quick Start `_

- `Docker `_

.. [#sumo] `SUMO home page `_

.. [#tianrang] `Tianrang Intelligence home page `_