Artificial Intelligence for Robotics¶
Course notes for CS 8803-O01 / CS 373 (Programming a Robotic Car), organized by lesson.
Lesson 1 — Localization¶
Histogram filters, probability, and grid-based localization.
- Localization
- Introduction
- Localization
- Total Probability
- Uniform Probability Quiz
- Uniform Distribution
- Generalized Uniform Distribution
- Probability After Sense
- Compute Sum
- Normalize Distribution
- pHit and pMiss
- Sum of Probabilities
- Sense Function
- Normalized Sense Function
- Test Sense Function
- Multiple Measurements
- Exact Motion
- Move Function
- Inexact Motion 1
- Inexact Motion 2
- Inexact Motion 3
- Inexact Move Function
- Limit Distribution Quiz
- Move Twice
- Move 1000
- Sense and Move
- Sense and Move 2
- Localization Summary
- Formal Definition of Probability 1
- Formal Definition of Probability 2
- Formal Definition of Probability 3
- Bayes’ Rule
- Cancer Test
- Theorem of Total Probability
- Coin Flip Quiz
- Two Coin Quiz
- Terms
- Resources
- Homework Assignment 1
- Office Hours Week 1
Lesson 2 — Kalman Filters¶
Continuous state estimation with Gaussians.
- Kalman Filters
- Introduction
- Tracking intro
- Gaussian Intro
- Variance Comparison
- Preferred Gaussian
- Evaluate Gaussian
- Maximize Gaussian
- Measurement and Motion 1
- Measurement and Motion 2
- Shifting the mean
- Predicting the Peak
- Parameter Update
- Parameter Update 2
- Separated Gaussians
- New Mean and Variance
- Gaussian Motion
- Predict Function
- Kalman Filter Code
- Kalman Prediction
- Kalman Filter Land
- Kalman Filter Prediction
- Another Prediction
- More Kalman Filters
- Kalman Filter Design
- Tracking Intro
- Gaussian Intro
- Variance Comparison
- Preferred Gaussian
- Evaluate Gaussian
- Maximize Gaussian
- Measurement and Motion 1
- Measurement and Motion 2
- Shifting the Mean
- Predicting the Peak
- Parameter Update
- Parameter Update 2
- Separated Gaussians
- Separated Gaussians 2
- New Mean and Variance
- Gaussian Motion
- Predict Function
- Kalman Filter Code
- Kalman Prediction
- Kalman Filter Land
- Kalman Filter Prediciton
- Another Prediction
- More Kalman Filters
- Kalman Filter Design
- Kalman Matrices
- Conclusion
- Kalman Filter Notes
- Homework Assignment 2
- Office Hours Week 2
Lesson 3 — Particle Filters¶
Monte Carlo localization and the robot class.
- Particle Filters
- Summary
- State Space
- Belief Modality
- Efficiency
- Exact or Approximate
- Particle Filters
- Using Robot Class
- Robot Class Details
- Moving Robot
- Add Noise
- Robot World
- Creating Particles
- Robot Particles
- Importance Weight
- Resampling
- Never Sampled 1
- Never Sampled 2
- Never Sampled 3
- New Particle
- Resampling Wheel
- Orientation 1
- Orientation 2
- Error
- You and Sebastian
- Filters
- 2012
- Preview
- Particle Filter Notes
- Homework Assignment 3
- Office Hours Week 3
- Conditional Particle Filters
Lesson 4 — Motion Planning¶
Grid search, A*, and dynamic programming.
Lesson 5 — PID Control¶
Path smoothing and robot control.
Lesson 6 — SLAM¶
Putting localization, planning, and control together.
- Putting It All Together
- Localization
- Planning
- PID
- Your Robot Car
- Your Robot Car
- Segmented CTE
- Fun with Parameters
- Wrap Up
- SLAM
- Is Localization Necessary
- Graph SLAM
- Implementing Constraints
- Adding Landmarks
- SLAM Quiz
- Matrix Modification
- Untouched Fields
- Omega and Xi
- Landmark Position
- Expand
- Introducing Noise
- Confident Measurements
- Implementing SLAM
- Congratulations
- SLAM Notes
- Homework Assignment 6
- Office Hours Week 6
Programming Exercises¶
Python templates live in code/ and are included from the lesson and homework pages.