Course Description

This course will cover basic principles for endowing mobile autonomous robots with perception, planning, and decision-making capabilities. Algorithmic approaches for robot perception, localization, and simultaneous localization and mapping; control of non-linear systems, learning-based control, and robot motion planning; introduction to methodologies for reasoning under uncertainty, e.g., (partially observable) Markov decision processes. Extensive use of the Robot Operating System (ROS) for demonstrations and hands-on activities. Prerequisites: CS 106A or equivalent, CME 100 or equivalent (for linear algebra), and CME 106 or equivalent (for probability theory).

Instructor

Prof. Marco Pavone

Course Assistants

Somrita Banerjee	Abhishek Cauligi	Boris Ivanovic
Mengxi Li	Joseph Lorenzetti

Meeting Times

Lectures meet on Tuesdays and Thursdays from 10:30am to 11:50am, Fridays from 10:00am to 11:20am. They will all be online. Friday lectures are optional for those enrolled in AA 174A. For all others, the homework will require material taught during the Friday lectures.

Sections are on Mondays and Wednesdays from 10:30am to 12:30pm, Mondays from 3:00pm to 5:00pm, Tuesdays from 4:00pm to 6:00pm, and Thursdays from 2:00pm to 4:00pm. They will all be online.

Prof. Pavone's office hours are on Tuesdays 1:00pm to 2:00pm and by appointment, all online.

CA office hours are on Tuesdays from 2:00pm to 4:00pm and Thursdays from 4:00pm to 6:00pm, all online.

Syllabus

The class syllabus can be found here.

Schedule

Subject to change. Lecture notes and scribe notes are updated below. We will try to have the lecture notes updated before the class.

Week	Topic	Lecture Slides	Lecture Notes	Sections
1	Course overview, mobile robot kinematics Introduction to the Robot Operating System (ROS) Friday: HW1 out	Lecture 1 Pre-knowledge quiz (solutions) Lecture 2 (live)	Lecture 1 Notes Lecture 2 Notes
2	Trajectory optimization Trajectory tracking & closed loop control Friday: Advanced methods for trajectory optimization	Lecture 3 Lecture 4 Lecture 5 (ex_1.py, ex_2.py)	Lecture 3 Notes Lecture 4 Notes Lecture 5 Notes	Section 1 Slides Section 1 Handout
3	Motion planning I: graph search methods Motion planning II: sampling-based methods Tuesday: HW1 due, HW2 out	Lecture 6 Lecture 7	Lecture 6 Notes Lecture 7 Notes	Section 2 Slides Section 2 Handout
4	Robotic sensors & introduction to computer vision Camera models & camera calibration Friday: Stereo vision Friday: HW2 due, HW3 out	Lecture 8 Lecture 9 Lecture 10	Lecture 8 Notes Lecture 9 Notes Lecture 10 Notes	Section 3 Slides Section 3 Handout
5	Image processing, feature detection & description Information extraction & classic visual recognition Friday: Modern robotic perception	Lecture 11 Lecture 12 Lecture 13	Lecture 11 Notes Lecture 12 Notes	Section 4 Slides Section 4 Handout
6	Intro to localization & filtering theory Parameteric filtering (KF, EKF, UKF) Friday: Nonparameteric filtering (PF) Tuesday: HW3 due, HW4 out	Lecture 14 Lecture 15 Lecture 16	Lecture 14 Notes Lecture 15 Notes Lecture 16 Notes	Section 5 Slides Section 5 Handout
7	EKF localization EKF SLAM Friday: Monte Carlo localization and particle filter SLAM Tuesday: Final project released	Lecture 17 Lecture 18 Lecture 19	Lecture 17 Notes Lecture 18 Notes	Section 6 Slides Section 6 Handout
8	Tuesday: No lecture Multi-sensor perception & sensor fusion (by Daniel Watzenig) Friday: No lecture	Lecture 20	Lecture 20 Notes	Section 7 Slides Section 7 Handout
9	Software for autonomous systems (by Daniel Watzenig) State machines Tuesday: HW4 due Friday: No lecture, Final project check-in due	Lecture 21 Lecture 22	Lecture 22 Notes
10	Decision making under uncertainty Reinforcement learning Friday: Final project demo	Lecture 23 Lecture 24	Lecture 23 Notes Lecture 24 Notes