In this special two-part episode with guest Stefano Marzani, the WW Tech Lead for Software Defined Vehicles from AWS, we discuss the importance of cloud technology as applied to vehicles. In this Part 2, we discuss Prototyping, Machine Learning/AI, and ADAS/Autonomous Driving.
In, Part 1, published previously (LINK) we meet Stefano, learn about his important role at AWS and cover the first two topics: Data and Compute.
Links referenced in the show:
Part 2 Chapters:
- 00:00 - Overview
- 00:40 - Topic 3: Prototyping in the cloud and “Environmental Parity”
- 02:20 - Running on Arm in the cloud and the vehicle
- 03:13 - Start software development in the cloud
- 03:55 - Difficulty of validation for automotive
- 04:49 - Empowering developers via the cloud
- 05:18 - Vehicle software will triple in the next five years
- 07:07 - “Shift left” to accelerate design cycles
- 07:54 - $40-50 Billion lost to automotive recalls
- 08:51 - Developing vehicle HMI with the cloud
- 10:57 - SDV can reduce cost of recalls
- 11:43 - Consolidation of ECUs
- 12:22 - Consolidating software effectively without bloat
- 13:45 - SOAFEE and standards for automotive
- 15:31 - Topic 4: Machine Learning and Analytics
- 16:45 - Data collection and annotation with SageMaker
- 17:22 - Other frameworks for ML in AWS
- 17:52 - In-vehicle validation of new models
- 18:30 - Finding edge cases
- 19:20 - Tools to manage ML workflows
- 21:55 - Topic 5: ADAS and Autonomous Driving
- 22:35 - Near-term benefits from ADAS
- 25:00 - Seeking edge cases
- 25:30 - Decomposing autonomous driving
- 27:30 - Collaboration between AWS and Sonatus
- 29:29 - Importance of using real ECUs
- 31:04 - Summary