Conference:  Transform X 2022

Authors: Neda Cvijetic, Russell Kaplan

2022-10-19

Neda Cvijetic, Senior Vice President of Autonomous Driving at Stellantis, joined Russell Kaplan, Scale’s Director of Engineering, for a fireside chat. The two will discussed the role of data diversity in building safer autonomy, and Cvijetic shared highlights from her career building autonomous vehicles at Tesla, NVIDIA, and now Stellantis, the parent company of Dodge, Fiat, and Chrysler.Cvijetic explained the opportunities that arise while building an entirely new infrastructure for training models for autonomous vehicles at a large automotive OEM, without the hindrance of having to support legacy systems. She also discussed how she plans to achieve Stellantis’s publicly shared goals around Level 3 autonomy in 2024. The Stellantis portfolio includes Jeep, meaning its systems will also handle off-road scenarios. Kaplan and Cvijetic also covered core paradigms in autonomous vehicles, from mapping to deploying to a million-vehicle fleet, to large language models, with more than a few surprising real-world anecdotes.Prior to Stellantis and NVIDIA, Cvijetic worked on autopilot and infotainment systems at Tesla, served on the adjunct faculty of Columbia University, and held senior research positions at NEC Labs America. She holds more than 20 U.S. patents.

Conference:  Transform X 2022

Authors: Dragomir Anguelov, Marco Pavone, Alex Kendall, Kate Park

2022-10-19

Experts discuss the challenges in incorporating machine learning into autonomous vehicles safely and effectively.

• Autonomous vehicles use multiple sensors to identify their surroundings, but face difficulties in identifying pedestrians, other vehicles, obstacles, and environmental conditions.
• Integrating complicated sensor suites, software, data management, and machine learning with engineering is a challenge.
• Collecting and labeling large amounts of data, integrating ML models with the rest of the self-driving stack, and improving the driver continuously are also challenges.
• Simulation plays a critical role in development.
• Different OEMs use unique approaches to leverage machine learning in their self-driving stack, with some using end-to-end learning and others preferring modular learning.
• Scaling to new environments quickly is a difficult challenge.