The anatomy of autonomous driving projects
summary
Continental's Autonomous Mobility team in Budapest runs a product-line for automated parking with ~250 engineers and thousands in the division, combining cameras, radar, lidar and ultrasonics into sensor fusion that produces top‑view occupancy grid maps and semantic labels; systems run at 25–30 Hz. A prototype once drove 110 km/h with a few lines of company code; current deliverables include an AI+classical hybrid that detects wheel stoppers, pedestrians and parking slots, supports hardware‑in‑the‑loop tests, CI/CT pipelines and staged continuous delivery.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 3:18Vision Zero and safer roads
- 7:11Self-driving levels and implications
- 10:00Parking scenario complexity
- 13:53Self-driving software architecture
- 14:56Sensing with AI and sensors
- 18:35AI for object detection
- 20:23Low vs high-speed driving
- 22:52Sensor fusion and safety redundancy
- 27:43Planning and system integration
- 30:15Scaling AI in automotive
- 33:50Agile teams in ADAS
- 37:47Next frontiers in AI
- 40:14Key takeaways
- 42:23Q&A
related talks
27mDriving Autonomous Vehicles Forward w/Real World Applications
31mDeep learning for self-driving vehicle technology
08mHuman–AI Collab for Reliable Automotive Control Systems
14mA new era of mobility : autonomous driving in 2021
05mInterview
19mChallenges & Opportunities for Building Crowdsourced Mapping Services for Autonomous Driving
40mBringing Automated Driving Software to the Production Vehicle
41mAutonomous vehicle development with Unity