ucla-mobility/OpenCDA

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1162 stars · 225 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

OpenCDA is an open-source Python framework for prototyping and evaluating full-stack cooperative driving automation (CDA) applications in a CARLA+SUMO co-simulation environment. It provides modular pipelines for perception, localization, planning, control, and V2X communication, along with benchmark scenarios and algorithms.

Use cases

• simulate cooperative adaptive cruise control and platooning scenarios
• evaluate V2X communication between connected automated vehicles
• benchmark cooperative perception algorithms in CARLA and SUMO
• prototype full-stack autonomous driving pipelines in Python
• run co-simulation of traffic flow and vehicle dynamics
• test CDA features for USDOT CARMA research

When to choose

• you need cooperative or V2X-enabled autonomous driving simulation, which single-vehicle simulators lack
• you want an all-Python full-stack CDA research platform with benchmarks
• you need CARLA and SUMO co-simulation integration

When to avoid

• you only need simple single-vehicle autonomous driving simulation without cooperation
• you need real-time or hardware-in-the-loop deployment rather than simulation
• you require a commercially licensed or fully production-grade driving stack

Facets

framework · maturity active

simulation machine-learning computer-vision robotics autonomous-vehicles simulation robotics python cross-platform autonomous-driving carla sumo v2x cooperative-driving platooning co-simulation connected-automated-vehicles perception planning-control research linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ucla-mobility/OpenCDA")