This seminar focuses on utilizing feedback control from pixels, exploring the integration of cameras into feedback loops using rigorous tools from systems theory. The course covers topics such as training correspondences, language of control, limitations of cameras, key point affordances, reinforcement learning, linear models, and image-based control. The teaching method involves discussing recent results and theoretical frameworks in the field. This course is intended for individuals interested in robotics, control theory, computer vision, and artificial intelligence.
Overview
Syllabus
Introduction
Welcome
Feedback control from pixels
Training correspondences
Language of control
Cameras
Limitations
Output feedback problem
Fundamental problem
General framework
State representation
Key point affordances
Key point base affordances
Modelbased policy search
Parameterizations
Reinforcement learning
Output feedback
RX models
State feedback
Linear models
Linear ARX models
A new problem
Carrots
Objective
Image
Image coordinates
Learning a model
Simple thought experiment
Action frame
Least squares
Linear map
Preimage
Closed loop performance
Next steps
Questions
Linear prediction
Robot representation
Diversity of tasks
Taught by
MIT Embodied Intelligence
Tags
Related Courses
Related articles
Reviews
3.0 rating, based on 1 Class Central review
-
The course I took was adequately informative and covered the essential concepts in a structured manner. The instructors were knowledgeable and provided clear explanations. However, there were limited opportunities for hands-on practice and interactive discussions, which could have enhanced the learning experience. The course materials were sufficient but lacked some depth in certain areas. Overall, it provided a foundational understanding of the subject, but additional resources or practical applications would have been beneficial for a more comprehensive learning experience.
