
Frank Dellaert | Factor Graphs for Perception and Action | Tartan SLAM Series
Published at : October 05, 2021
Presentation by Frank Dellaert titled, "Factor Graphs for Perception and Action" as part of the Tartan SLAM Series.
Series overviews and links can be found on our webpage: https://theairlab.org/tartanslamseries2/
Outline:
00:00 - Welcome and Intro
01:50 - Introduction to Factor Graphs for Perception and Action
04:34 - Outline
05:31 - Why Factor Graphs?
08:11 - Factor Graphs for Perception
19:13 - Trajectory Optimization
30:14 - Optimal Control
43:45 - Model-predictive Control
49:21 - Conclusion
50:14 - Q&A
50:30 - Q&A: Book recommendation for factor graphs
51:53 - Q&A: IMU preintegration
53:42 - Q&A: Constraint optimization
56:17 - Q&A: Robot-world interaction and deliberate contact
59:41 - Q&A: GPU factor graph acceleration
01:02:02 - Q&A: Temporal difference in loop-closure detection with factor graph
01:04:32 - Q&A: Balance of different factors in the factor graph
01:06:54 - Ending
Series overviews and links can be found on our webpage: https://theairlab.org/tartanslamseries2/
Outline:
00:00 - Welcome and Intro
01:50 - Introduction to Factor Graphs for Perception and Action
04:34 - Outline
05:31 - Why Factor Graphs?
08:11 - Factor Graphs for Perception
19:13 - Trajectory Optimization
30:14 - Optimal Control
43:45 - Model-predictive Control
49:21 - Conclusion
50:14 - Q&A
50:30 - Q&A: Book recommendation for factor graphs
51:53 - Q&A: IMU preintegration
53:42 - Q&A: Constraint optimization
56:17 - Q&A: Robot-world interaction and deliberate contact
59:41 - Q&A: GPU factor graph acceleration
01:02:02 - Q&A: Temporal difference in loop-closure detection with factor graph
01:04:32 - Q&A: Balance of different factors in the factor graph
01:06:54 - Ending

RoboticsSLAMLocalization