Chapter
Scaling Lateral's Approach to Static and Dynamic Worlds
1:29:03 - 1:31:42 (02:39)
Lateral's approach for reinforcement learning on the world will scale to handle both the static and dynamic elements. The future plan is to perform reinforcement learning on the world once all the cars come in.
Clips
The speaker explains that using LIDAR camera fusion is the best way to build a highly reliable perfect localizer for autonomous vehicles, which can work for over 100,000 miles.
1:29:03 - 1:30:20 (01:16)
LIDAR Camera Fusion
Summary
The speaker explains that using LIDAR camera fusion is the best way to build a highly reliable perfect localizer for autonomous vehicles, which can work for over 100,000 miles.
ChapterScaling Lateral's Approach to Static and Dynamic Worlds
EpisodeGeorge Hotz: Comma.ai, OpenPilot, and Autonomous Vehicles
PodcastLex Fridman Podcast
The approach towards lateral handling will scale completely to both static and dynamic elements.
1:30:20 - 1:31:42 (01:22)
Autonomous Driving
Summary
The approach towards lateral handling will scale completely to both static and dynamic elements. The next step is to incorporate reinforcement learning to predict the behaviour of these elements in autonomous driving.