Chapter
Learning to Drive with Data
1:02:30 - 1:06:46 (04:15)
The use of data and machine learning could enable cars to automatically do lane changes and learn appropriate driving policies from good drivers.
Clips
The ability for self-driving cars to make lane changes automatically can be learned from data by creating a labeled dataset that shows when to make the lane change, allowing the car to make a change that is appropriate for the situation.
1:02:30 - 1:05:50 (03:20)
Self-Driving Cars
Summary
The ability for self-driving cars to make lane changes automatically can be learned from data by creating a labeled dataset that shows when to make the lane change, allowing the car to make a change that is appropriate for the situation. By studying the behavior of good drivers, we can create a set of policies that can become universally applicable.
ChapterLearning to Drive with Data
EpisodeGeorge Hotz: Comma.ai, OpenPilot, and Autonomous Vehicles
PodcastLex Fridman Podcast
The main advice for Tesla is to put driver monitoring in their cars, but for cars that don't already have this feature, it is difficult to improve the autonomy.
1:05:50 - 1:06:46 (00:55)
Tesla
Summary
The main advice for Tesla is to put driver monitoring in their cars, but for cars that don't already have this feature, it is difficult to improve the autonomy.