COMPE696: AI for Unmanned Systems [Spring 2025]

Instructor	Dr. Junfei Xie https://smile.sdsu.edu/
Zoom link	https://SDSU.zoom.us/j/84630330018   (This link serves as a hub for office hours, lecture recordings, and various other course-related activities.)
Office Hours	Tuesdays: 8:40am - 9:20am & 11:00am - 11:30am  Thursdays: 8:40am - 9:20am (in-person, online or by appointment)
Office Location	SDSU Engineering Building E-403B
Contact Info.	Email: jxie4@sdsu.edu
Prerequisites	COMPE510
TA Info	Haomeng Zhang (Email: hzhang3986@sdsu.edu)
TA's Office Hours	By appointment

*For any questions related to programming assignments, contact the TA first. For other questions, contact the instructor.

**A brief version of the syllabus can be downloaded here. 

Course Description:

In recent years, there has been a notable trend towards integrating unmanned systems (US) into the fabric of future smart cities. Artificial intelligence (AI) plays a pivotal role in enabling US to execute a diverse array of tasks with enhanced efficiency, precision, and autonomy. Through AI algorithms, US can autonomously control movement, process data from sensors, and make informed decisions based on collected information.

This course offers a comprehensive introduction to US and the application of AI in this domain. Key topics covered include fundamentals of AI, deep learning, US navigation and control, sensing and estimation techniques, mission and path planning, US communication and computing, autonomy, AI applications to US, especially unmanned aerial systems, and more. Students will engage in practical exercises to implement various techniques and develop US applications using AI methodologies.

Course Learning Objectives:

Following this course, students will be able to:

• Compare the common types and applications of US.
• Explain basic AI concepts and techniques.
• Compare the various types of sensors typically used in US and choose suitable sensor fusion methods.
• Explain the common methods used for planning the path for US.
• Produce programs to model, simulate, and control US.
• Design and implement AI-based unmanned system applications, as well as analyze and present the results.

References

• Quan, Quan. Introduction to multicopter design and control. Singapore: Springer, 2017. [Quan]
• Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT press, 2016. [Goodfellow]
• Sutton, Richard S., and Andrew G. Barto. Reinforcement learning: An introduction. MIT press, 2018. [Sutton]
• LaValle, Steven M. Planning algorithms. Cambridge university press, 2006. [LaValle]
• Choset, H., Lynch, K. M., Hutchinson, S., Kantor, G. A., & Burgard, W. Principles of robot motion: theory, algorithms, and implementations. MIT press, 2005. [Choset]
• Russell, Stuart J., and Peter Norvig. Artificial intelligence: a modern approach. London, 2010. [Russell]

 

Grade Scale:

A = 94.0-100.0

A- = 90.0-93.9

B+ = 87.0-89.9

B = 84.0-86.9

B- = 80.0-83.9

C+ = 77.0-79.9

C = 74.0-76.9

C- = 70.0-73.9

D+ = 67.0-69.9

D = 64.0-66.9

D- = 60.0-63.9

F = 59.9 and below

I = Did Not Complete

 

Grade Distribution and General Comments:

• 2 exams: 16% + 16%  = 32%
• Participation in in-class discussion: 5%
• Programming Assignments: 8% + 8% + 8% = 24%
• Research Paper Presentation: 5% 
• Term Project: 34%
  • Proposal presentation: 3%
  • Mid-term presentation: 5%
  • Final presentation: 8%
  • Final report: 18%
  • Teaser video (optional): up to 3% as a bonus
  • CPS-IoT Week Competition (optional): up to 3% as a bonus

*Grades for this course may be curved, taking into account overall class performance.

 

Exams:

There will be two short exams during the semester. There will be no make-up exams, so please plan accordingly. 

 

Participation in in-class discussion:

Students are required to attend every class in person and participate in in-class discussions. 

 

Programming Assignments:

There will also be 3 individual programming assignments based on Python, which must successfully compile and properly execute on your own PC/MAC/Workstation/Laptop. I might selectively ask a few students to demonstrate their program.

I will not accept late assignments so please make sure you turn in your assignments on time. Soft copies of the assignments should be turned in through Canvas.

 

Research Paper Presentation:

Each student will individually develop and deliver a presentation on an assigned research paper. Through a signup sheet, you will select a research paper that interests you and choose a presentation date. Slots will be allocated on a first-come, first-served basis.

 

Term Projects:

Students will form teams of up to three members to collaboratively work on a term project. Each team will deliver three presentations: a project proposal, a mid-term progress update, and a final presentation. Additionally, a comprehensive final report summarizing the project outcomes is required. To enhance the presentation of key results, creating a teaser video is highly encouraged. Teams that submit a teaser video will earn a bonus of up to 3% of the total grade. 

Students are also highly encouraged to participate in the 4th CPS-IoT Week Student Design Competition on Networked Computing on the Edge. Teams participating in the competition will earn a bonus of up to 3% of the total grade. 

 

Late Assignment Policy:

Assignments are considered late if they are submitted after the due date and time. Late assignments are NOT acceptable.

 

General Comments:

** NOTE: Some of the documents (handouts, slides, or assignment instructions) may have accessibility issues. If you encounter such issues, please contact me by email. 

 

Tentative Lecture Schedule for January

1/21	T	Explore syllabus
1/23	Th	Introduction, Machine Learning Overview	Read [Russell, Chapter 1]
1/28	T	Machine Learning Overview, Deep Neural Networks		Presentation Topic & Date due
1/30	Th	Deep Neural Networks	Read [Goodfellow, Chapters 6-7]

 

Tentative Lecture Schedule for February

2/4	T	Deep Neural Networks		Last day to drop class (2/3)!
2/6	Th	Deep Neural Networks
2/11	T	Convolutional Neural Networks		Presenter: Dhruv Donga Title: "VL-MFL: UAV Visual Localization Based on Multisource Image Feature Learning"
2/13	Th	Convolutional Neural Networks, Reinforcement Learning		Presenter: Nilesh Singh Title: "Intrusion Detection for Unmanned Aerial Vehicles Security: A Tiny Machine Learning Model"
2/18	T	Reinforcement Learning		Presenter: Jacob Anderson Title: "Energy Consumption Optimization of UAV-Assisted Traffic Monitoring Scheme With Tiny Reinforcement Learning"
2/20	Th	UAV Overview		Presenter: Jiya Rathi Title: "A Lightweight Reinforcement-Learning-Based Real-Time Path-Planning Method for Unmanned Aerial Vehicles" Programming Assignment 1 due
2/25	T	Quadrotor Dynamics		Presenter: Abraham Carranza Title: "Learning Target-Aware Vision Transformers for Real-Time UAV Tracking"
2/27	Th	Quadrotor Dynamics		Presenter: Cagla Ipek Kocal Title: "A Deep-Learning Method Based on the Multistage Fusion of Radar and Camera in UAV Obstacle Avoidance"

 

Tentative Lecture Schedule for March

3/4	T	Term Project Proposal Presentation		Presenter: Parshav Pagaria Title: "AoI-Aware Energy-Efficient SFC in UAV-Aided Smart Agriculture Using Asynchronous Federated Learning"
3/6	Th	Exam 1		CPS-IoT Competition Registration due (3/9)!
3/11	T	Flight Control		Presenter: Ushasri Badinidi Title: "Transformer-Based Reinforcement Learning for Scalable Multi-UAV Area Coverage"
3/13	Th	Flight Control		Presenter: Arkajit Dutta Title: "An Efficient Matching Game Approach to Association Formation in UAV-Enabled Hierarchical Distributed Learning"
3/18	T	Flight Control		Presenter: Omar Ahmed Title: "UAV-Assisted Covert Federated Learning Over mmWave Massive MIMO"
3/20	Th	Sensing and Estimation		Presenter: Larry Garcia Title: "Dense Multiagent Reinforcement Learning Aided Multi-UAV Information Coverage for Vehicular Networks"
3/25	T	Sensing and Estimation		Presenter: Eric Rosas Title: "Color Consistency of UAV Imagery Using Multichannel CNN-Based Image-to-Image Regression and Residual Learning"
3/27	Th	Sensing and Estimation		Presenter: Rosa Miranda Title: "Deep Reinforcement Learning Enabled Multi-UAV Scheduling for Disaster Data Collection With Time-Varying Value" Programming Assignment 2 due

 

Tentative Lecture Schedule for April

4/1	T	Spring Recess. No class!
4/3	Th	Spring Recess. No class!
4/8	T	Term Project Mid-term Presentation		Presenter: Dhruv Shah Title: "Federated Learning-Based Task Offloading in a UAV-Aided Cloud Computing Mobile Network"
4/10	Th	Sensing and Estimation		Presenter: Oluchi Nzerem Title: "Bias-Compensation Augmentation Learning for Semantic Segmentation in UAV Networks"
4/15	T	Path Planning		Presenter: Phu Tran Title: "Cooperative Tracking of Quadrotor UAVs Using Parallel Optimal Learning Control"
4/17	Th	Path Planning		Presenter: Ty Runner Title: "Deep Reinforcement Learning for Energy-Efficient Data Dissemination Through UAV Networks"
4/22	T	Path Planning		Presenter: Disha Santosh Sawant Title: "Centroid-Guided Target-Driven Topology Control Method for UAV Ad-Hoc Networks Based on Tiny Deep Reinforcement Learning Algorithm"
4/24	Th	Path Planning		Presenter: Nihal Azeez Title: "Game of Drones: Intelligent Online Decision Making of Multi-UAV Confrontation" CPS-IoT Competition Final Product due (4/27)!
4/29	T	Advanced topics on UAV		Presenter: Surender Varma Title: "Mobility Management for Cellular-Connected UAVs: Model-Based Versus Learning-Based Approaches for Service Availability"

 

Tentative Lecture Schedule for May

5/1	Th	Term Project Final Presentation		Programming Assignment 3 due
5/6	T	Term Project Final Presentation
5/8	Th	Exam 2		Term project final report due

---

This course will use the Canvas Learning Management System instead of Blackboard. To access your course log in at canvas.sdsu.edu, and sign in using your SDSUid.

Note: You are responsible for adjusting your notification settings in such a way that you receive ALL announcements regarding this class. All Canvas email notifications will be delivered to your SDSU email address. You can add additional email addresses and sign up for text/mobile app notifications via the settings in your Canvas Profile, and then adjust your notifications in the Notifications Tab. Canvas notifications are system wide and cannot be adjusted by course. Click here to view a step-by-step guide to add additional notification and contact methods. 

If you have technical issues with Canvas, please contact the SDSU Canvas 24/7 support line at 619.483.0632.

---