Brain

Autoware

Open-source autonomous driving software stack maintained by the Autoware Foundation. Built on ROS 2, widely used in research and commercial autonomous vehicle projects. Tier IV is the primary commercial contributor.

Framework · Maturity: Production · Open source

Machine-readable surfaces

Markdown mirror: /brains/autoware.md
RSS feed: /brains/autoware/feed.xml
JSON-LD: embedded in this page’s head
REST API: /v1/brains/f9404dff-1532-45bf-a26e-195800ed27a3
Revision history: /brains/autoware/history
Data documentation: /data
Query this programmatically: Deploy MCP

Architecture

ROS 2-based, modular perception-planning-control stack

Key facts

Stack scope (primary Autoware Foundation): The Autoware Foundation overview says Autoware is an open-source autonomous driving project built on ROS, with modules for sensing, localization, perception, planning, control, vehicle interface, and map data. It says the project is designed toward curb-to-curb Level 4 driving. 'World's leading' on that page is foundation marketing, not an independent ranking (Autoware Foundation)
Core, Universe, and stated scale (primary Autoware Foundation): The same overview splits Autoware Core (ODD functionality under a stricter review and release process) from Autoware Universe (extra packages for research, including a Robo-Taxi ODD not yet in Core). The page displays counters of 30 vehicle types, 20 countries, and 500 companies. Treat those counters as foundation-displayed figures, not a census Rover verified (Autoware Foundation)
ROS 2 documentation: Autoware's documentation site provides installation and usage documentation for its open source autonomous driving software stack and current ROS 2 based releases (Autoware documentation)
Autoware Foundation project home: The Autoware Foundation presents Autoware as an open source autonomous driving project intended to cover sensing, localization, perception, planning, control, and vehicle interfaces (Autoware Foundation)
Open-source stack: The Autoware Foundation describes Autoware as an open-source autonomous driving software stack (Autoware Foundation).
ROS 2 foundation: Autoware's official documentation describes the stack's ROS 2-based architecture for autonomous driving applications (Autoware documentation).
Modular autonomy: Autoware's documentation organizes the stack around modules for sensing, localization, perception, planning, control, and vehicle integration (Autoware documentation).
End to end evolution: The Autoware Foundation's E2E report describes a direction from a traditional robotics stack toward a data-centric, AI-based autonomous driving stack that incrementally replaces hand-coded modules with neural networks (Autoware E2E report).

Developed by (1)

Tier IV

Common questions

What is Autoware?

Who developed Autoware?

Autoware is credited to Tier IV on the DEPLOY registry. Each developer attribution is verified via primary sources.

Is Autoware open source?

Yes. Autoware is recorded as open-source / open-weights on the DEPLOY registry, meaning model weights or source code are publicly available.

What type of AI is Autoware?

Autoware is a framework, built on a ROS 2-based, modular perception-planning-control stack architecture on the DEPLOY registry.

What is Autoware's maturity stage?

Autoware is at the production stage on the DEPLOY maturity ladder. Production stage means high-volume sustained commercial operations.

Which robots run on Autoware?

No robot models on the DEPLOY registry are recorded as running Autoware. DEPLOY wires brain-to-model connections only when the wiring is verifiable from primary sources; absence may reflect pre-deployment or unverified manufacturer claims.

Sources (5)

Methodology: Verified · 5 sources (no primary) · last reviewed 2026-10-09

Verification posture

Verified

Low confidence

Review state

Stable

Last reviewed 2026-10-09

Maturity + lifecycle

Maturity stage: production

Sources by quality tier

5: unclassified; Unclassified source

The framework is documented at /methodology. Corrections at /corrections. Reviewer: DEPLOY editorial team.

Canonical ID f9404dff-1532-45bf-a26e-195800ed27a3