Driving Autonomous Vehicles Forward w/Real World Applications
summary
A deployed urban AV stack that combines LiDAR, cameras and radar with a fifth‑generation "wayo driver" claims 20 million autonomous miles and 20 billion simulation miles of training. Perception, semantic understanding, prediction and planning run on onboard AI compute; real incidents handled include red‑light runners, rogue motorbikes and cyclists splitting lanes. Products operating in Chandler, Phoenix and San Francisco plus long‑haul trucking in Texas show the system working in dense intersections, narrow gaps and inclement weather under a published safety framework.
AI-written summary · about
chapters (15)
- 0:00Autonomous vehicle tech intro
- 1:10Why autonomous driving matters
- 3:18Waymo’s decade of safety
- 4:42Waymo Driver hardware
- 6:28Software and AI backbone
- 8:33Safety and performance at scale
- 11:04Real-world ride-hailing
- 12:32Waymo One in action
- 14:57Urban driving challenges
- 16:00Cyclist and pedestrian safety
- 18:31Autonomous trucking platform
- 19:37Building customer trust
- 21:22Q&A on intentions
- 23:50Sensor redundancy in L4
- 24:54Scaling evaluation with miles
related talks
23mA new way forward for self driving cars
37mBuckle Up: How Waymo Uses Data to Create a Safer Driving Experience
26mAutonomy – The Era of Vehicles Without Drivers
31mDeep learning for self-driving vehicle technology
51mTeaching Autonomous Vehicles to Drive with Visible and Non-Visible Light Simulation
21mAutonomous driving and handling the handover
16m↑↑↓↓← → ← → Autonomous
41mAutonomous vehicle development with Unity