CMPT 882

Special Topics in Artificial Intelligence: Robotic Decision Making

Lecture times and location:

• MWF 12:30-13:20, Academic Quadrangle 5030

Instructor

• Mo Chen
  SFU Burnaby, TASC 1 8225
  mochen@cs.sfu.ca
  Office hours: Thursdays 15:00 - 16:00

Lecture notes

• Jan. 4: Introduction
• Jan. 7: Linear Algebra Review
• Jan. 9: Linear Algebra Review II
• Jan. 11: Linear Systems I
• Jan. 14: Linear Systems II
• Jan. 16: Nonlinear Systems I
• Jan. 18: Nonlinear Systems II
• Jan. 21: Lyapunov Stability
• Jan. 23: Guest Lecture by Ian Mitchell, UBC
• Jan. 25: Numerical Methods for Solving ODEs I
• Jan. 28: Numerical Methods for Solving ODEs II
• Jan. 30: Convex Optimization I
• Feb. 1: Convex Optimization II
• Feb. 4: Convex Optimization III
• Feb. 6: Nonlinear Optimization
• Feb. 8: Optimal Control I
• Feb. 11: Optimal Control II. Demo code
• Feb. 13: Optimal Control III
• Feb. 15 Dynamic Programming I
• Feb. 25 Dynamic Programming II
• Feb. 27 Continuous-Time LQR and Introduction to Reachability
• Mar. 1 Hamilton-Jacobi Reachability I
• Mar. 4 Hamilton-Jacobi Reachability II
• Mar. 6 Sampling-Based Motion Planning
• Mar. 8 Sensors and Regression Overview
• Mar. 11 Regression
• Mar. 13 Neural Networks and Markov Decision Processes
• Mar. 15 Imitation Learning
• Mar. 18 Introduction to Reinforcement Learning
• Mar. 20 Reinforcement Learning II
• Mar. 22 Reinforcement Learning III
• Mar. 25 Bayes' Filter
• Mar. 27 Kalman Filter
• Mar. 29 EKF SLAM

Assignments

• Assignment 1: Due Feb. 4. Solutions
• Assignment 2: Due Mar. 18. Solutions
• Assignment 3: Due Apr. 20. Solutions

Instructor's Objectives

This course provides an overview of robotic planning and decision making algorithms, with a focus on mobile robots. Following a brief introduction to robotic systems, the course will cover popular computational methods and algorithms for solving planning and decision making problems. An emphasis will be placed on using state-of-the-art computational tools in practical settings. In addition, students will gain exposure to results and challenges in recent research. Topics include modeling of robotic systems, motion planning, optimal control, optimization, robotic safety, machine learning, robotic perception. Applications include unmanned aerial vehicles, self-driving cars, and household robots.

Grading

• Homework: 40%
• Project: 60%

Project options

• Thoroughly understand and critically evaluate 3 to 5 papers in an area covered in this course.

• Reproduce the results of 1 to 2 papers in an area covered in this course, and suggest or make improvements

• Mini Research project related to an area covered in this course.

• Other: please consult with instructor

Project timeline

• Consultation throughout the term
• Proposal (1-2 paragraphs) Due Feb. 18
• Presentations in the last three classes
• Project report (4 pages maximum) due Apr. 20

Project topics

• Dynamical systems
• Optimization
• Optimal control
• Machine learning in robotics
• Localization and mapping

Recommended textbooks

• R. Siegwart, I. R. Nourbakhsh, and D. Scaramuzza, Introduction to Autonomous Mobile Robots. The MIT Press, 2011, 9780262015356.
• S. S. Sastry, Nonlinear Systems: Analysis, Stability, and Control. Springer-Verlag, 1999, 9780387985138
• S. M. LaValle, Planning Algorithms. Cambridge University Press, 2006, 9780521862059.
• S. Boyd and L. Vandenberghe, Convex Optimization. Cambridge University Press, 2008, 9780521833783.
• D. P. Bertsekas, Dynamic Programming and Optimal Control. Athena Scientific, 2017, 1886529434.

Academic Honesty Statement

Academic honesty plays a key role in our efforts to maintain a high standard of academic excellence and integrity. Students are advised that ALL acts of intellectual dishonesty will be handled in accordance with the SFU Academic Honesty and Student Conduct Policies ( http://www.sfu.ca/policies/gazette/student.html ).