COMP 765: Intelligent Robotics, Winter 2019
News
- Jan 30, 2019: Seminar paper sign-up is today! Also Assignment 1 has just been released and check the updated schedule for precise dates of presentations to the end of class.
- Jan 23, 2019: Please sign-up for Piazza which we're going to use to discuss assignments and papers from here on: piazza.com/mcgill.ca/spring2019/comp765
- Jan 21, 2019: Slide links up to lecture 6 have just been refreshed (including the pptx versions). So please download fresh if you'd like the latest!
- Jan 21, 2019: The reading list has been posted just below. Sign-up happening soon!
- Jan 9, 2019: Please fill out the Welcome Survey which will be used to tailor the seminar to interests of the group this year.
- Jan 7, 2019: Welcome to the course!
Overview
COMP 765 is a research seminar on Intelligent Robotics: the intersection of rocket science and the latest machine learning on-board systems that can sense the world and act upon it. The class will begin with lectures on definitional problems and algorithms in robotics. We will then transition to mixed student-lead and instructor-lead discussions of recent developments in research and in practice. The emphasis is on algorithms, probabilistic reasoning, learning to improve behaviors using data, and decision making under uncertainty, as opposed to electromechanical systems design. We will broadly cover the following areas:
- Sensing and Estimation: robots (like humans) sense their surroundings through limited, imprecise measurements. What algorithms are needed to give robots the ability to reason accurately about the world?
- Control and Planning: how can we compute the state-(in)dependent commands that can bring a robotic system from its current state to a desired state, when models of the robot and its environment are available?
- Decision Making Under Uncertainty: how can the same problems be solved when our model knowledge is missing or imprecise? In particular, what motion strategies best complement our probabilistic estimators, for example by gathering useful information or acting conservatively?
- Learning for Robotics: how can we use data to learn models for a robot's continuous time series data in case these are not known in advance?
- Heirarchical Decision Making: how can tasks with longer time-scale or for larger target systems be factored productively into multiple levels of reasoning?
- Human-robot Interaction: how can we make programming robots easy? Can robots learn from the way humans perform and describe tasks? Can we achieve super-human performance? Can robots teach humans how to best perform or reason about tasks?
Seminar Format
The class will be a mix between lecture-style teaching and student-lead discussion. In each unit, I will lead things off by presenting standard material given in textbooks, covering the theory and analysis with basic examples. We will then transition to the state-of-the-art in the area, with the end of the unit featuring presentations of important recent papers following the PROPONENT, OPPONENT, REPRODUCER format to be described in class.
The [READING LIST], along with the schedule and assigned presenters is located here. This link is edittable. Please feel free to offer your peers trades, the spreadsheet can be edited using this link.
Assignments
- Assignment 1: Robot Localization (on github)
- Assignment 2: Learning and Controlling Dynamic Systems (on github)
Schedule
Date | Topics | Slides | References |
---|---|---|---|
Jan 7 |
Introduction Motivation and course sylabus |
(pdf) (pptx) | PR Chapter 1. |
Jan 9 |
Spatial Representations Kinematics and Dynamics |
(pdf) (pptx) | PR Chapter 5. |
Jan 14 and 16 |
Intro to Estimation Sensing uncertainty, Bayesian filters, Kalman basics |
(pdf) (pptx) | PR Chapter 3 |
Jan 21 |
EKF and Particle Filters Importance sampling, analysis, efficient resampling methods |
(pdf) (pptx) | PR Chapter 4 |
Jan 23 |
Modern Localization and Mapping Rao-Blackwellized Particle Filtering, large-scale sparse graph optimizers, mapping an entire city. |
(pdf) (pptx) | PR Section 3 |
Jan 28 |
Example Seminar Presentation Dave and Travis as Proponent and Opponent present ORBSLAM. |
(pdf) (pptx) | ORBSLAM Paper |
Jan 30 |
Gaussian Processes For Robot Learning Bayesian regression, random processes, GP formulation, inference and parameter learning. |
(pdf) (pptx) | Slides from Uni Freiburg course GP text first 3 chapters |
Feb 4 |
Estimation Papers Day Student-lead seminars for 3 papers |
Send me your slides when ready so I can post them. | As listed in the reading list |
Feb 6 and Feb 11 |
Robotic Control Intro Control theory, ODE background, PID, MDPs, dynamic programming |
(pdf) (pptx) | PLAN text, selected ideas from Chapter 15 |
Feb 13 |
Trajectory Optimization LQR, DDP, policy search |
(pdf) (pptx) (recording) | Underactuated robotics text, Chapter 9 - LQR |
Feb 18 |
Control Papers Day Student-lead seminars for 3 papers |
Send me your slides when ready so I can post them. | As listed in the reading list |
Feb 20 |
Multi-task and Robust Control Contextual and transfer learning overview, domain randomization, policy adjustment. |
(pdf) (pptx) | Behavior Adaptation |
Feb 25 |
Learning to learn Meta learning, generalized policies and value functions. |
(pdf) (pptx) | Dropout as a Bayesian Approximation |
Feb 27 |
Transfer Learning Papers Day Student-lead seminars for 3 papers |
Student Preseter's Slides | As listed in the reading list |
March 4 and 6 |
McGill Reading Week No class. Take a rest! |
||
March 11 |
Human Robot Interaction Intro HRI overview, imitation learning, behavior cloning, DAGGER. |
(pdf) (pptx) | An Invitation to Imitation |
March 13 |
Inverse Reinforcement Learning MaxEnt IRL, Interactive Reward Learning, outperforming the demonstrator. |
||
March 18 |
HRI Papers Day Student-lead seminars for 4 papers |
Student presenter's slides | As listed in the reading list |
March 20 |
Planning Introduction Planning flavors: geometric, symbolic, discrete, continuous. Dynamic programming and graph solutions. Start of RRT. |
(pdf) (pptx) | PLAN text first 3 chapters |
March 25 |
Sampling-based and uncertain planning. Finish RRT details, intuitive proofs. Probabilistic completeness. POMDP overview. |
(pdf) (pptx) | PLAN text RRT material PR chapters 15 and 16 |
March 27 |
Planning under uncertainty POMDP details, approximate planning, Bayesian optimization. |
(pdf) | Tutorial on Bayesian Optimization |
April 1 |
Planning Papers Day Student-lead seminars for 3 papers |
Send me your slides when ready so I can post them. | As listed in the reading list |
April 3 |
Multi-robot systems Collaborative manipulation, swarm robotics, decision making under communication and coordination constraints. |
(pdf) (pptx) | |
April 8 |
Multi-robot Papers Day Student-lead seminars for 3 papers |
Send me your slides when ready so I can post them. | As listed in the reading list |
April 10 |
Project presentation day Will be in an extended class or poster session. More details as the date approaches. |
Send me your slides when ready so I can post them. |
Marking scheme
- 3 assignments worth 10% each = 30%
- 3 in-class presentations worth 10% each = 30%
- The first as a PROPONENT, who presents the main content of the paper and gives its positive aspects
- The second as an OPPONENT, who focuses on the weaknesses or limitations of the paper, perhaps drawing on aspects where the related work is stronger
- The third as a REPRODUCER, who has tried the authors code or data (or re-implemented aspects of the paper) and reports on this experience
- 1 final project written report worth 40%
Recommended, but optional, textbooks
- Probabilistic Robotics, by Thrun, Fox, and Burgard (PR in the outline)
- Planning Algorithms, by Lavalle (PLAN in outline)
- Gaussian Processes for Machine Learning, by Rasmussen and Williams (GPforML in the outline)
- Computational Principles of Mobile Robotics, 2nd edition, by Dudek and Jenkin
- Robotics, Vision, and Control, by Corke
- Introduction to Autonomous Mobile Robots, by Siegwart, Nourbakhsh, Scaramuzza
Related courses
- Pieter Abbeel's course
- Sebastian Thrun's Udacity course
- Related sections from Stephen Boyd's linear systems course
- Related sections from Russ Tedrake's underactuated robotics course