BWSI Course - Autonomous Air Vehicle (UAV)

Rapidly expanding unmanned aerial vehicle (UAV) technology has enabled a number of new application areas. The growth in UAV development is evident in the popularity of First Person View (FPV) drone racing and interest from companies, like Amazon and others, to develop fully autonomous aerial delivery vehicles. As UAV technologies mature, they open new and exciting areas for potential research. This summer Beaver Works will offer students the opportunity to explore some of these new areas of research and tondesign their own autonomous capabilities for UAVs. The students will work in teams to develop algorithms for deployment to a commercial quadrotor, the DJI Tello drone. They will use the various open-source libraries, and custom algorithms that they will develop. The summer course will culminate in a competition at which the students will apply the knowledge gained from the four-week program’s projects and lectures to a series of racing challenges.


This program consists of two components: an online course from February to May open to all interested students and a four-week summer program at MIT from July for a small group of students. The online component gives students a background in the course material and provides a solid mathematical foundation that will be critical when completing the more advanced topics of the summer course. The skills the students develop during the summer will be demonstrated in labs and at the final event. Students will demonstrate basic implementations of control and autonomy after each unit of instruction. These lessons will build upon previous instruction to enable students to develop algorithms so that a quadrotor can autonomously navigate a UAV racecourse designed for the summer program. 

Online Course

The online component for the Autonomous Air Vehicle Racing course will contain important introductory material that will provide students with the background required to successfully complete the four-week summer course. In addition to the introductory material, the online course will include more advanced, quadrotor-specific material so that students can begin to explore problems specific to autonomous aerial vehicles.


Introduction and Prerequisites

  • Introduction to quadrotors
  • Linear algebra
  • Basics of matrix mathematics
  • Introduction to probability and statistics
  • Computer programming fundamentals

Autonomous Aerial Vehicles

  • Flight geometry
  • Actuators and control
  • State estimation
  • Sensing
  • Basic control theory
  • Computer vision
  • Visual motion estimation

Summer Course

The four-week summer program will be structured to provide the students with projects and hands-on exercises. The program will apply and expand upon the online course material, leading to multiple competitive team challenges in autonomous UAV control. Each day in the course will consist of a mix of lectures and hands-on projects to reinforce and apply the material. A team of experienced MIT researchers will provide the lectures each day, covering material that reviews UAV and autonomy fundamentals and expanding on advanced topic areas in the lecturer’s expertise. Hands-on projects will enable the students to apply each lecture, working toward a capability for autonomous UAV racing by using the provided Tello drone and associated experimentation equipment. In addition, the course is lining up guest lecturers from among leading researchers in the computer science, autonomy, and air vehicle academic and corporate communities to provide the students with emerging trends in these fields. Upon completion of the four-week course, the students will have developed an understanding of autonomous systems development, including controls, flight dynamics, navigation, and computer vision.


The course curriculum is new this year and is still in development at this time, but the current plan extends over three weeks of instruction and hands-on practice and one week of team challenges, culminating in the final UAV racing challenge. The detailed topics for each week are listed below:


Week 1: Flight

  • Quadrotor design
  • Quadrotor dynamics
  • Quadrotor components

Week 2: Vision

  • Image formation
  • Edge detection
  • Image filtering
  • Object detection

Week 3: Control

  • Control systems
  • State estimation
  • Navigation and planning

Week 4: Racing Challenges


The final week of the course will focus on hands-on team projects in autonomous UAVs and racing challenges, leveraging the lessons learned from the first three weeks of the course.