# LucasAlegre/sumo-rl

Reinforcement Learning environments for Traffic Signal Control with SUMO. Compatible with Gymnasium, PettingZoo, and popular RL libraries.

Repository: https://github.com/LucasAlegre/sumo-rl
Canonical: https://ross.abutalabs.com/products/sumo-rl
Homepage: https://lucasalegre.github.io/sumo-rl
Language: Python
License: MIT
License Family: permissive
Topics: sumo, rl-algorithms, reinforcement-learning, gym, traffic-signal-control, gym-env, pettingzoo, machine-learning, deep-reinforcement-learning, python, gymnasium
Last push: 2026-03-08T13:23:47+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 71, release rhythm 8, longevity 100
- inputs: {"age_days": 2823, "days_push": 178, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1071, forks 265 (observed 2026-08-28T04:03:28.128351+00:00)

## What it is
SUMO-RL is a Python library that provides Gymnasium and PettingZoo-compatible reinforcement learning environments for traffic signal control using the SUMO traffic simulator. It supports single-agent and multiagent RL with customizable state and reward definitions, and integrates with libraries like stable-baselines3 and RLlib.

## Use cases
- train RL agents to control traffic lights in SUMO simulations
- run multiagent reinforcement learning experiments on traffic networks
- benchmark traffic signal control algorithms with Gymnasium environments
- customize observation and reward functions for traffic signal agents
- use stable-baselines3 or RLlib to optimize traffic congestion
- simulate and evaluate adaptive traffic signal timing policies

## When to choose
- you need RL environments for traffic signal control built on SUMO
- you want Gymnasium or PettingZoo API compatibility for traffic experiments
- you need multiagent RL support for intersection networks
- you want easily customizable state and reward definitions for traffic agents

## When to avoid
- you need traffic simulation without reinforcement learning
- you require a different traffic simulator than SUMO
- you need real-world traffic signal control rather than simulation

## Facets
- artifact type: library
- maturity: active
- function: reinforcement-learning, simulation, machine-learning
- domain: reinforcement-learning, simulation, machine-learning
- platform: python, windows, cross-platform
- tags: sumo, traffic-signal-control, gymnasium, pettingzoo, multiagent-rl, traci, traffic-simulation, transportation, linux, macos

## Member repositories
- LucasAlegre/sumo-rl (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:28.128351+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:53:44.997464+00:00, confidence not recorded.
  - readme: https://github.com/LucasAlegre/sumo-rl (fetched 2026-08-28T04:03:28.128351+00:00, sha 9360cd3b92f2)
  - homepage: https://lucasalegre.github.io/sumo-rl (fetched 2026-08-29T12:56:09.768185+00:00, sha 1d09a41a7061)
  - registry_pypi: https://pypi.org/pypi/sumo-rl/json (fetched 2026-08-29T12:56:09.771840+00:00, sha 368be2a1d775)
- Data as of 2026-08-30T08:39:29.467469+00:00.