Lecture 3: Visual Navigation for Flying Robots (Dr. Jürgen Sturm)

Описание

Probabilistic Models and State Estimation

Topics covered:
- Bayesian Probability Theory
- Bayes Filter
- Normal Distribution
- Kalman Filter
- Examples

Course website:
http://vision.in.tum.de/teaching/ss2013/visnav2013

Speaker: Dr. Jürgen Sturm
http://vision.in.tum.de/members/sturmju

Slide Titles:
00:00:06 Scientific Research
00:01:40 Flow of Publications
00:03:56 Important Conferences in Robotics and Computer Vision
00:05:11 Perception
00:07:43 Optical Illusions
00:08:06 Models in Human Perception
00:08:18 State Estimation
00:10:59 Models and State Estimation
00:11:58 Deterministic Sensor Model
00:12:49 Deterministic Motion Model
00:13:24 Probabilistic Robotics
00:16:13 Agenda for Today
00:16:22 The Axioms of Probability Theory
00:17:19 A Closer Look at Axiom 3
00:17:44 Discrete Random Variables
00:18:49 Continuous Random Variables
00:20:16 Proper Distributions Sum To One
00:20:31 Joint and Conditional Probabilities
00:22:28 Conditional Independence
00:23:38 Marginalization
00:24:51 Law of Total Probability
00:25:22 Expected Value of a Random Variable
00:27:06 Covariance of a Random Variable
00:27:35 Estimation from Data
00:28:07 The State Estimation Problem
00:28:48 Causal vs. Diagnostic Reasoning
00:29:58 Bayes Formula
00:30:52 Normalization
00:31:54 Bayes Rule with Background Knowledge
00:32:02 Example: Sensor Measurement
00:35:01 Combining Evidence
00:36:05 Recursive Bayesian Updates
00:37:47 Example: Second Measurement
00:39:10 Actions (Motions)
00:40:21 Typical Actions
00:42:00 Action Models
00:42:17 Example: Take-Off
00:43:46 Integrating the Outcome of Actions
00:46:10 Markov Chain
00:47:53 Markov Assumption
00:49:44 Bayes Filter
00:52:28 Example: Localization
00:57:15 Bayes Filter - Summary
00:58:49 Kalman Filter
00:59:12 Normal Distribution
01:02:31 Properties of Normal Distributions
01:04:24 Linear Process Model
01:06:39 Linear Observations
01:07:33 Kalman Filter: Variables and Dimensions
01:08:23 From Bayes Filter to Kalman Filter
01:09:43 Kalman Filter: Algorithm
01:12:27 Nonlinear Dynamical Systems
01:13:28 Taylor Expansion
01:14:26 Extended Kalman Filter
01:15:01 Kalman Filter
01:15:46 Kalman Filter: Example
01:24:59 Lessonos Learned Today