Главная
Study mode:
on
1
Intro
2
Motivation
3
Autonomous Driving Pipeline
4
End-to-End Learning
5
Challenges
6
Talk Outline
7
Guardian Angel
8
Parallel Autonomy:Architecture
9
Parallel Autonomy: Hardware
10
Shared # Binary Control
11
Possible Approaches
12
Autonomous Modes
13
Related Work End-co-End Learning
14
Learning a Steering Distribution
15
Discrete Action Learning
16
Multimodal Distributions
17
Advantages of this approach
18
Dataset Collection
19
Discrete to Continuous
20
Variational Bayes Mixture Models
21
Bounds for Parallel Autonomy
22
Why Care About Uncertainty?
23
Bayesian Deep Learning
24
End to End Steering Control
25
Integrating Uncertainty Estimation
26
A Bayesian Outlook on End to End Control
27
Elementwise Dropout for Uncertainty
28
Spatial Dropout for Uncertainty
29
Training Results
30
Summary
Description:
Explore an invited talk from the GPU Technology Conference (GTC) 2018 that delves into learning steering bounds for parallel autonomy and handling ambiguity in end-to-end driving. Discover the latest methodologies for training end-to-end systems in autonomous vehicles, focusing on the challenges of integrating decision-making capabilities beyond reactionary control. Examine the concept of parallel autonomy, its architecture, and hardware implementation. Investigate approaches to learning steering distributions, including discrete action learning and multimodal distributions. Gain insights into dataset collection, Bayesian deep learning, and uncertainty estimation in end-to-end steering control. Analyze the advantages and limitations of these approaches in the context of autonomous driving pipelines and higher-level decision making.

Learning Steering for Parallel Autonomy

Alexander Amini
Add to list
0:00 / 0:00