Teaching

This is a Course-based Undergraduate Research Experience (CURE) course aimed to provide hands-on research experience in a lab setting. 

This course revolves around building a small-scale autonomous racing car that can navigate a circular track reliably and safely despite unexpected environment changes (e.g., different illumination or previously unseen obstacles) and mechanical degradations (e.g., slippery surface or depleted battery). Specifically, these 1/16th to 1/10th-scale racing cars are based on commercial RC hardware and equipped with a camera or a lidar sensor. Our deep learning-based software runs on a Linux system, which is deployed on an NVidia Jetson platform, and sends commands to an electronic speed controller that actuates the car’s motor and servos.

To be successful, this end-to-end robotic development requires diverse expertise from mechanical, electrical, and computing domains. Undergraduate students should expect to be assigned tasks that match their initial skills and also develop new competencies along the way. In the long term, autonomous racing cars will serve as a physical platform for evaluating and demonstrating the latest research of the TEA (Trustworthy Engineered Autonomy) Lab in predicting and validating the safety of autonomous systems. The skills and knowledge developed in our lab can be applied to a broad range of other autonomous systems and robots. 

Under the supervision of Dr. Ivan Ruchkin and experienced student mentors, undergraduate students in this course will (i) assemble and configure the mechanical and electrical hardware of racing cars, (ii) program and train perception and control components for racing cars, (iii) assist comprehensive literature search and review, (iv) participate in weekly engineering meetings and research discussions. Depending on the circumstances, undergraduate students will also have an opportunity to contribute to research publications as co-authors.

Credits: 0 – 3.

Instructors

• Ivan Ruchkin, faculty
• Ao Wang, graduate mentor
• Sam Jhong, graduate mentor
• Nathan Norohna, graduate mentor 
• Lorant Domokos, undergraduate mentor
• Yuang Geng, graduate meta-mentor
• Zhenjiang Mao, graduate meta-mentor

This course covers the fundamentals of data structures and algorithms, including lists, queues, stacks, divide-and-conquer, dynamic programming, trees, tables, graphs, and recursive techniques. It also investigates the role of specific data structures in electrical engineering applications.

Programming has become an essential component of virtually every aspect of Engineering.  However, writing correct and efficient programs requires an understanding of the underlying foundations, including implementation, manipulation, and analysis of structured data, understanding how algorithms are built on top of such data, and navigating trade-offs between program performance and resource constraints. This course covers the foundations of programming specifically targeted toward Electrical Engineers. It will cover the implementation and use of data structures in solving programming problems, including queues, trees, tables, and graphs.  Algorithm development including recursive techniques will be discussed, and several algorithmic design techniques (greedy, dynamic programming, branch and bound) will be presented.  The course will include excursions illustrating the use of these techniques in a variety of Electrical Engineering domains.

Credits: 3. 

The complete syllabus can be found here.

This course will teach mathematical and algorithmic techniques and tools for building safe autonomous systems, from unmanned drones to self-driving cars to medical devices. 

The course is split into three modules: 

1. Systems: how to understand and formally represent a complex autonomous system? Mathematical formalisms for dynamical, probabilistic, and hybrid systems.
2. Autonomy: how to enable autonomous operation in a challenging environment? Data-driven components, probabilistic graphical models, neural networks, errors/robustness of learned models, learned perception/control.
3. Safety: how to ensure that an autonomous system behaves safely? Logical specifications of safety, verification of system models and learned components, safe planning and control, temporal and confidence monitoring.

Credits: 3.

See the syllabus for details. Feel free to reach out if you have any questions about the course or your participation in it.

If you are interested in the teaching materials, please reach out as well.