Toward Total Scene Understanding for Autonomous Driving-Drago Anguelov (Waymo)
Вставка
- Опубліковано 14 гру 2024
- Ben Taskar Distinguished Memorial Lecture
Title: Toward Total Scene Understanding for Autonomous Driving
Speaker: Drago Anguelov (Waymo)
Host: Anat Caspi
Date: January 25, 2024
Abstract: Machine learning has proven to be a key ingredient in building a performant and scalable Autonomous Vehicle stack, spanning key capabilities such as perception, behavior prediction, planning and simulation and evaluation. I will describe recent Waymo research on performant ML models and architectures that help us handle the variety and complexity of real world environments. I will also discuss how progress in building Autonomous Driving agents can impact people with disabilities and cover some current open questions about how to further enhance embodied AI agent capabilities with ML.
Bio: Drago joined Waymo in 2018 to lead the Research team, which focuses on pushing the state of the art in autonomous driving using machine learning. Earlier in his career he spent eight years at Google; first working on 3D vision and pose estimation for StreetView, and later leading a research team which developed computer vision systems for annotating Google Photos. The team also invented popular methods such as the Inception neural network architecture, and the SSD detector, which helped win the Imagenet 2014 Classification and Detection challenges. Prior to joining Waymo, Drago led the 3D Perception team at Zoox.
This video is closed captioned.
A version of this video with ASL interpretation is available here: • [ASL] Toward Total Sce...
Currently watching.
If 30 years from now all all vehicles are AVs would this software be the same?
The software will evolve as new research comes in. For example the neural net architectures shown in this video some of them weren't even't invented when google started its self driving project.
The software is constantly changing. It will be vastly different in thirty years.