Secure Autonomous and Cyber-Physical Systems

Instructor: Prof. Sibin Mohan, Oregon State University

CS/ECE 599 | Winter 2022 Term | MW 2:00 - 3:50 PM PT | BAT 150

This course will focus on the state-of-the-art on security for autonomous and cyber-physical systems. The goal is to provide a deep and broad understanding of the design of such systems as they interact with the real world. Autonomous cars, unmanned aerial/ground/water vehicles, robots, etc. are examples of systems that we intend to explore.


Smart City with Autonomous Systems


We will cover topics related to:

Hands-on exercises will explore well known, open-source simulation and control platforms such as:

The course will also include an exploration of the state-of-the-art research papers in these domains.

Prerequisites: students should be familiar with one or more modern programming languages such as Python, C++, etc. The hands-on machine problems will involve the setting up of open-source simulation frameworks (detailed instructions will be provided).


Course Details

[* Note: Please message ahead of time and set up alternate slot in case of conflicts for office hours.]

Communication

We will use Canvas for all course related communications.

Target Audience

Graduate students in CS, ECE, Robotics, AI and even other departments who want to explore topics related to the security, privacy and ethics of autonomous and cyber-physical systems.

Covid

If you suspect you have been exposed to someone with Covid, or are feeling sick yourself, please let me know and I will provide a Zoom link for you to attend remotely. And get some rest!

Grading Criteria

In-class participation 5 %
MP1 [V2X/VEINS] 20 %
MP2 [UAV Security/Gazebo] 8 %
MP3 [V2X Attack/Leaderboard/VEINS] 37 %
Paper Reading/Critique 30 %

Conversion from numerical to letter grade is as follows:

95 - 100 A
90 - 94 A-
87 - 89 B+
83 - 86 B
80 - 82 B-
77 - 79 C+
73 - 76 C
70 - 72 C-
67 - 69 D+
63 - 66 D
60 - 62 D-
Below 60: F

In Class Participation

I encourage you to ask questions and ask often. Participate in discussions as well. Take notes since that helps with the learning.

Schedule

Week Date Topic Links
1 Jan 3, 2022
Intro to Autonomy, Cyber-Physical Systems and Real-Time
Systems. Discussion on course syllabus and requirements
Lecture Slides
Jan 5, 2022
Machine Problem MP I-A: Introduction to VEINS/FMD
Due Date: Jan 14, 2022 [11:59 PM PT]
MP I-A
Jan 5, 2022
Functional Design of Autonomous Systems and Sensors

Additional Information:
Lecture Slides
2 Jan 10, 2022 Lecture Slides
Jan 12, 2022
Localization and Sensor Fusion

"A Novel Side-Channel in Real-Time Schedulers"
by C. Y. Chen, A. Ghassami, S. Mohan, R. Bobba,
R. Pellizzoni and N. Kiyavash, IEEE RTAS 2019.

Additional Information:
Lecture Slides

Scheduleak Slides
3 Jan 17, 2022
No class [Dr. Martin Luther King Day!]
Jan 19, 2022
Kalman Filtering

"TaskShuffler: A Schedule Randomization Protocol
for Obfuscation Against Timing Inference Attacks
in Real-Time Systems
"
by Man-Ki Yoon, Sibin Mohan, Chien-Ying Chen
and Lui Sha, IEEE RTAS 2016.

Additional Information:
Lecture Slides

Obfuscation Slides
Jan 22, 2022
Machine Problem MP I-B: V2X Misbehavior Detectors
Due Date: Feb 04, 2022 [11:59 PM PT]
MP I-B
4 Jan 24, 2022
SLAM

"Guaranteed Physical Security with
Restart-Based Design for Cyber-Physical Systems
"
by F. Abdi, C. Y. Chen, M. Hasan, S. Mohan
and M. Caccamo, ACM/IEEE ICCPS 2018.

Additional Information:
Lecture Slides

ReSecure Slides
Jan 26, 2022 Lecture Slides
Jan 29, 2022
Paper reading list announced
Papers List
5 Jan 31, 2022
Discussion on Security Defenses
See Lecture 7 Slides
Feb 02, 2022
Paper Presentation 1: Opeyemi Ajibuwa
Paper: Too Good to Be Safe: Tricking Lane Detection
in Autonomous Driving with Crafted Perturbations

by Jing et al., USENIX Security 2020.

Paper Presentation 2: Kyle Trevis
Paper: Overfitting, Robustness, and Malicious Algorithms:
A Study of Potential Causes of Privacy Risk
in Machine Learning

by Yeom et al., Journal of Computer Security, 2020.
Paper Slides




Paper Slides
6 Feb 07, 2022
Paper Presentation 1: Nora Basha
Paper: Wireless Attacks on Aircraft Instrument
Landing Systems

by Sathaye et al., USENIX Security 2019.

Paper Presentation 2: Jason Pratama
Paper: Automated Cross-Platform Reverse Engineering
of CAN Bus Commands From Mobile Apps

by Wen et al., NDSS 2020.
Paper Slides




Paper Slides
Feb 09, 2022
Paper Presentation 1: Brandon Ellis
Paper: Adaptive Adversarial Videos on Roadside Billboards:
Dynamically Modifying Trajectories of Autonomous Vehicles

by Patel et al., IROS 2019.
[local pdf]

Paper Presentation 2: Matthew Phillips
Paper: Robust Physical-World Attacks on Deep Learning
Visual Classification

by Eykholt et al., CVPR 2018.
Paper Slides





Paper Slides
Feb 12, 2022
Machine Problem MP II-A: PX4 & uORB Introduction
Due Date: Feb 18, 2022 [11:59 PM PT]
MP II-A
7 Feb 14, 2022
Paper Presentation 1: Shahab Nikkhoo
Paper: Darts: Deceiving autonomous cars with toxic sign
by Sitawarin et al., ArXiv 2018.

Paper Presentation 2: Opeyemi Ajibuwa
Paper: Secure Pose Estimation for Autonomous Vehicles
under Cyber Attacks

by Liu et al., IV 2019.
Paper Slides



Paper Slides
Feb 16, 2022
Paper Presentation 1: Matthew Phillips
Paper: DIAT: Data Integrity Attestation for Resilient
Collaboration of Autonomous Systems

by Abera et al., NDSS 2019.

Paper Presentation 2: Nora Basha
Paper: A Confidence Range Tolerant Misbehavior
Detection Approach

by Kamel et al., IEEE WCNC 2018.
Paper Slides




Paper Slides
8 Feb 21, 2022
Paper Presentation 1: Jason Pratama
Paper: MCC-EKF for Autonomous Car Security
by Singandhupe et al., IRC 2020.

Paper Presentation 2: Brandon Ellis
Paper: Standard detectors aren't (currently) fooled
by physical adversarial stop signs

by Lu et al., ArXiv 2017.
Paper Slides



Paper Slides
Feb 23, 2022
Paper Presentation 1: Kyle Trevis
Paper: Privacy-Preserving Tampering Detection
in Automotive Systems

by Roman et al., Electronics 2021.

Paper Presentation 2: Shahab Nikkhoo
Paper: Noise Matters: Using Sensor and Process Noise
Fingerprint to Detect Stealthy Cyber Attacks and
Authenticate sensors in CPS

by Ahmed et al., ACSAC 2018.
Paper Slides




Paper Slides
9 Feb 28, 2022
Paper Presentation 1: Matthew Phillips
Paper: Privacy in Autonomous Vehicles
by Glancy et al., Santa Clara Law Review, 2012.


Paper Presentation 2: Opeyemi Ajibuwa
Paper: Privacy and Integrity Considerations
in Hyperconnected Autonomous Vehicles

by Karnouskos et al., IEEE 2018.
Paper Slides




Paper Slides
Feb 28, 2022
Machine Problem MP II-B: PX4 MiTM attacks [Extra Credit]
Due Date: Mar. 07, 2022 [11:59 PM PT]
MP II-B
Mar. 01, 2022
Machine Problem MP III: V2X Misbehavior Detection Contest
Due Date: Mar. 12, 2022 [11:59 PM PT]
MP III
Mar 02, 2022
Paper Presentation 1: Kyle Trevis
Paper: Privacy and performance trade-off
in cyber-physical systems

by Zhang et al., IEEE Network 2016.

Paper Presentation 2: Nora Basha
Paper: The epistemic opacity of autonomous systems
and the ethical consequences

by Héder, AI & Soc 2020.
Paper Slides




Paper Slides
10 Mar 07, 2022
Paper Presentation 1: Shahab Nikkhoo
Paper: (De)constructing Ethics for Autonomous Cars:
A Case Study of Ethics Pen-Testing towards
‘AI for the Common Good'

by Berendt et al., IRIE 2020.

Paper Presentation 2: Brandon Ellis
Paper: Avoiding the Intrinsic Unfairness
of the Trolley Problem

by Holstein et al., FairWare 2018.
Paper Slides




Paper Slides
Mar 09, 2022
Paper Presentation 1: Jason Pratama
Paper: Will the Driver Seat Ever Be Empty?
by Fraichard, INRIA Research Report 2014.

Class Wrap Up
Paper Slides

Policies

Late Submissions for MPs

Submission/due dates are announced along with the MP descriptions.

You have a total of 3 "grace" days to use during the term. You can use those to submit whatever assignment(s) you want late. E.g. you can turn in one assignment 2 days late, the other 1 day late. Or turn in 3 assignments, each 1 day late. If you want to use these late days, indicate so in the comment box when you submit it on Canvas.

You start losing 20% of the points per day after the grace days you use. E.g., if the assignment is due on Tuesday, and you turn it in on Wednesday without saying anything, you lose 20 points. If you use 1 grace day, and turn it in on Wednesday, you don't lose any points; but if you turn it in on Thursday, your final grade will be Your Earned Grade - 20. On Friday, it will be Your Earned Grade - 40.

An assignment can be at most three days late: so in our example, if you turn in the assignment Saturday, the grade is 0.

The point of this policy is to balance the need to be fair to those who turn in their assignments on time, with the need to recognize that sometimes, stuff happens and you just can't do all that's expected of you. Why have deadlines at all? Because almost everything in life does, and this is a small-stakes environment to practice that. Also, we can't grade everything at once.

Collaboration and the Honor Code

You are encouraged to work together. You may discuss the MPs with other people to understand the problem and reach a solution. However, each student/group must write down the solution independently, without referring to written notes from others. Hence, you must understand the solution well enough to discuss it yourself. In addition, each student/group must explicitly mention the names of the people with whom they collaborated. If I suspect cheating, I might have the student(s) come and answer questions in my office. If my suspicions are confirmed, I will refer the student to the disciplinary committee.

The purpose of problem sets in this class is to help you think about the material, not just give us the right answers. You are encouraged to use online resources for learning more about the material covered in class; however, you should not look for or use found solutions to questions in the problem sets. Specifically, you must not look at any code that has been created to solve the assignment, including solutions found on the internet to questions in the problem sets, code created by a student in a previous class or code created by a current classmate. (Though frankly, I seriously doubt you'll find anything, I regularly change the assignments).

Familiarize yourself with the standards set forth in the OSU Code of Student Conduct section 4.2. If there is any question about whether an act constitutes academic misconduct, it is your responsibility to seek clarification and approval from the instructor prior to acting. When in doubt, ask, or don't do it.

Absence Policy

If you miss a class, for whatever reason, you are still responsible for that class' material. Go through the lectures, consult the scribe notes, come to office hours (in that order), and talk to your fellow students, to learn the material.

Grade Review Policy

You have one week after a grade is released to ask me questions about it and seek a correction. After the week has passed, the grade is finalized. When releasing a grade, I also post comments where appropriate, explaining where you lost points, made mistakes, etc.

In case you are seeking a correction to the grade, you need a specific reason: e.g., your reasoning on MP 1.a was essentially correct but your written explanation, you now realize, was ambiguous; my comments say that your model had errors and I couldn't run it, but you can run it fine on your machine; etc. I cannot accommodate general requests that are a variation on "I think my grade was too low". Why do you think that? Based on our discussion, I might increase the grade, decrease it, or leave it as is.

Of course, you can stop by anytime during the term to ask questions about any part of the material. The one-week deadline applies to grade-specific questions.

COVID-Related Topics

The university's COVID-19 Safety and Success website serves as the primary communication channel for the Oregon State community regarding COVID-19 logistics and classroom guidelines.

Face Coverings

The university has a requirement to use a face covering when in indoor spaces in order to contribute to the health and safety of the OSU community during the ongoing COVID-19 pandemic. Accordingly, you are expected to use a face covering when attending class, including when you speak. I will do the same.

Acceptable face coverings include:

The following are not acceptable face coverings: face coverings that incorporate a valve to facilitate easy exhalation, mesh masks, lace masks, face shields or other coverings with openings, holes, or visible gaps in the design, material or vents.

If you don't have your own, face coverings are available at the library, Memorial Union or Student Health Services. All CoE buildings also have masks available.

Requests for exemption from wearing a face covering in indoor spaces require approval through the disability accommodation process. Please contact Disability Access Services (DAS) if you are seeking accommodation to the Policy based on disability. I encourage you to review the policy yourself, to understand the university's expectations around the use of face coverings outside the classroom, OSU's Vaccination requirement, and OSU's isolation and quarantine requirements.

Students with Disabiities

Accommodations for students with disabilities are determined and approved by Disability Access Services(DAS). If you, as a student, feel that you need accommodations but have not obtained approval please contact DAS immediately at 541-737-4098 or at http://ds.oregonstate.edu. DAS notifies students and faculty members of approved academic accommodations and coordinates implementation of those accommodations. While not required, students and faculty members are encouraged to discuss details of the implementation of individual accommodations.

The definition of disability is not limited to physical disability. OSU’s Nondiscrimination on the Basis of Disability policy states:

"...no qualified person shall, solely by reason of disability, be denied access to, participation in, or the benefits of, any program or activity provided by the University. Each qualified person shall receive the reasonable accommodations needed to ensure equal access to employment, educational opportunities, programs and activities in the most integrated setting feasible."

Reach Out for Success

University students encounter setbacks from time to time. If you encounter difficulties and need assistance, it.s important to reach out. Consider discussing the situation with an instructor or academic advisor. Learn about resources that assist with wellness and academic success at oregonstate.edu/ReachOut. If you are in immediate crisis, please contact the Crisis Text Line by texting OREGON to 741-741 or call the National Suicide Prevention Lifeline at 1-800-273-TALK (8255). If you have a disability that requires special accommodation, please get in touch with the University's Disability Access Services. The definition of disability is not limited to physical disability. They will work with the instructor to make sure this class works for you: http://ds.oregonstate.edu