Brain
Didi's autonomous driving unit developing robotaxi technology for the Chinese market. Raised $1.5B total funding. Operating robotaxi services in multiple Chinese cities.
Foundation model · Maturity: Commercial · Closed
Machine-readable surfaces
- Markdown mirror: /brains/didi-driver.md
- RSS feed: /brains/didi-driver/feed.xml
- JSON-LD: embedded in this page’s head
- REST API: /v1/brains/4a1cb613-10a8-4e36-9347-b69210c29674
- Revision history: /brains/didi-driver/history
- Data documentation: /data
- Query this programmatically: Deploy MCP
Architecture
Full-stack autonomous driving system
Key facts
- R2 driverless trials (primary PR Newswire / DiDi)
- DiDi Autonomous Driving's PR Newswire release says it began fully driverless service trials with its next-generation Robotaxi R2. Treat trial start and R2 designation as company-stated (PR Newswire / DiDi)
- Multi-city trial framing (secondary IEV China)
- Secondary coverage of the same R2 program describes driverless trials spanning Chinese cities and Abu Dhabi with GAC Aion L4 vehicles. Prefer DiDi's primary release for facts; treat city list as reported (IEV China)
- Level 4 autonomous driving
- DiDi says its autonomous driving unit develops Level 4 technology and has pursued public-road testing permissions in multiple Chinese cities and California (DiDi Autonomous Driving).
- Full-stack autonomy capabilities
- DiDi lists HD maps, positioning, environment perception, route planning, simulation, infrastructure, commercial operation, and mobility data among its autonomous-driving capabilities (DiDi Autonomous Driving).
- Full-stack autonomy scope
- DiDi says its autonomous-driving technology covers perception, prediction, decision-making, control, simulation, and machine-learning platforms (DiDi intelligent driving).
- Sensor and compute stack
- DiDi identifies lidar, cameras, 4D millimeter-wave radar, infrared and acoustic sensors, and its Orca three-domain central computing platform as parts of its autonomous-driving technology (DiDi intelligent driving).
- R2 sensor count
- DiDi's R2 robotaxi release says the next-generation vehicle uses 33 sensors, triple-domain fusion computing, and layered safety redundancy (DiDi R2 driverless trials).
Common questions
What is Didi Autonomous Driving System?
Didi's autonomous driving unit developing robotaxi technology for the Chinese market. Raised $1.5B total funding. Operating robotaxi services in multiple Chinese cities.
Is Didi Autonomous Driving System open source?
No. Didi Autonomous Driving System is recorded as proprietary on the DEPLOY registry. Model weights and source are not publicly available.
What type of AI is Didi Autonomous Driving System?
Didi Autonomous Driving System is a foundation model, built on a Full-stack autonomous driving system architecture on the DEPLOY registry.
What is Didi Autonomous Driving System's maturity stage?
Didi Autonomous Driving System is at the commercial stage on the DEPLOY maturity ladder. Commercial stage means production-grade deployments are operating at named customer sites.
Which robots run on Didi Autonomous Driving System?
No robot models on the DEPLOY registry are recorded as running Didi Autonomous Driving System. 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 (3)
- https://www.prnewswire.com/news-releases/didi-autonomous-driving-begins-fully-driverless-service-trials-with-next-generation-robotaxi-r2-302864755.html
- https://www.didiglobal.com/science/intelligent-driving
- https://ievchina.com/ai-mobility/didi-robotaxi-r2-2026-driverless-trials-beijing-guangzhou-abu-dhabi-gac-aion-l4/
Methodology: Verified · 3 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: commercial
Sources by quality tier
- 3
- unclassified
- Unclassified source
The framework is documented at /methodology. Corrections at /corrections. Reviewer: DEPLOY editorial team.
Methodology surface for Didi Autonomous Driving System.Canonical ID 4a1cb613-10a8-4e36-9347-b69210c29674