Главная
Study mode:
on
1
Introduction
2
Welcome
3
Feedback control from pixels
4
Training correspondences
5
Language of control
6
Cameras
7
Limitations
8
Output feedback problem
9
Fundamental problem
10
General framework
11
State representation
12
Key point affordances
13
Key point base affordances
14
Modelbased policy search
15
Parameterizations
16
Reinforcement learning
17
Output feedback
18
RX models
19
State feedback
20
Linear models
21
Linear ARX models
22
A new problem
23
Carrots
24
Objective
25
Image
26
Image coordinates
27
Learning a model
28
Simple thought experiment
29
Action frame
30
Least squares
31
Linear map
32
Preimage
33
Closed loop performance
34
Next steps
35
Questions
36
Linear prediction
37
Robot representation
38
Diversity of tasks
Description:
Explore feedback control using camera inputs in this MIT Embodied Intelligence Seminar featuring Russ Tedrake, Toyota Professor at MIT. Delve into the challenges of integrating control theory with visual feedback, examining recent advancements in reinforcement learning and imitation learning. Learn about attempts to bridge the gap between camera-based control and traditional systems theory, with Tedrake presenting recent results and small steps towards this goal. Gain insights into topics such as training correspondences, key point affordances, model-based policy search, and linear ARX models. Discover how these concepts apply to robotics, including discussions on robot representation and the diversity of tasks in feedback control from pixels.

Feedback Control from Pixels

Massachusetts Institute of Technology
Add to list
0:00 / 0:00