The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) System Design for Autonomous Vehicles with Drago Anguelov - #454
Feb 8, 2021
Drago Anguelov, Distinguished Scientist and Head of Research at Waymo, dives into the exciting world of autonomous vehicles. He shares insights on the significant advancements in AV technology and the focus on achieving level 4 driving capabilities. Drago breaks down critical ML use cases like Perception and Simulation, showcasing how Waymo’s fully autonomous systems are performing in Phoenix. He also discusses the socioeconomic impacts of self-driving cars and the potential influence of AV systems on future enterprise machine learning.
AI Snips
Chapters
Transcript
Episode notes
Drago's Journey to Waymo
- Drago Anguelov, head of Waymo Research, initially resisted working on self-driving cars.
- He joined Waymo after realizing deep learning's potential in 3D perception and prediction.
Waymo's Holistic Approach
- Waymo's autonomous driving system blends custom hardware, in-house sensors, and sophisticated software.
- This approach prioritizes a modular design supporting cars and trucks.
Scaling to New Cities
- Scaling autonomous driving to new cities involves technical and regulatory challenges.
- Adaptability and minimizing human oversight are key for broader deployment.