The future of transportation is uncrewed and ultimately autonomous. Before these systems can be fielded, they will need to go through a risk mitigation and acceptance process that is managed by professionals in and out of the government. This course is designed to give the students the skills required to evaluate autonomous aviation systems during this Development, Test & Evaluation, Verification and Validation (DTEVV) process. The course will cover the basic concepts and challenges faced by the DTEVV community when evaluating systems that contain Autonomous/AI functionality. It will examine control methods that are used by the DTEVV community. The course will provide practical experience in evaluating these systems through formal methods analysis, reinforcement learning, and programming to perform perception tasks required within the DTEVV community. This course will give the student a solid understanding and the skills needed to perform DTEVV on autonomous/AI systems in aviation.

Provides a comprehensive and rigorous introduction to computer vision principles and their application in the Development, Test & Evaluation, and Verification & Validation (DTEVV) of autonomous systems. Students will explore the entire computer vision pipeline, from the physics of image formation to the implementation and validation of state-of-the-art deep learning models. The course is grounded in the theoretical foundations of machine learning and deep learning, while also emphasizing the practical challenges of building robust and reliable systems. A significant portion of the course will be dedicated to hands-on labs and a final project where students will develop, train, and rigorously evaluate a computer vision model for an object tracking task, simulating real-world DTEVV scenarios by analyzing performance against both nominal and disturbed data.