CSE - Multi-Drone Swarming
The DroneResponse Lab is building the next generation of autonomous drone teams to support urgent, high-impact missions, including search and rescue, disaster assessment, and emergency recovery. We develop advanced multi-vehicle UAV systems that can operate in real-world, high-stakes environments, adapting to uncertainty, navigating complex terrain, and supporting human decision-makers when every second counts. Our work combines autonomous decision-making, real-time anomaly detection, computer vision, and human-centered situational awareness, pushing the frontier of intelligent aerial systems. For a quick overview of our Drone Response project please see: https://youtu.be/DyKqxkesgg0?si=OnNHl0Y6dSzn_Ap-
We are seeking an undergraduate research assistant to help with experiments and data collection for a novel visualization and supervision tool for multi-drone swarming operations. The project focuses on how to display the positions and movements of multiple drones in a realistic scene view, and your work will center on validating and evaluating the system using both simulation data and real-world flight recordings. Tasks may include assisting with field-work drone data collection, running experiment trials, organizing datasets, and analyzing accuracy and performance results. The student will work closely with our research team. Particularly strong contributions will lead to co-authoring an upcoming research paper.
This position is a great opportunity for students interested in drones, robotics, data analysis, and computer vision. Some experience with Python or quantitative analysis is helpful but not required — curiosity, reliability, and attention to detail matter most.
DroneResponse is a Software and Systems engineering research lab. Your project will include some of the following in our lab:
- Practice software architecture and design thinking: You'll go beyond just writing code, thinking about how the system is structured, how different parts of the autonomy pipeline connect, and how to make those components modular, reusable, and testable. You’ll help sketch out and implement core elements like decision-making modules, data flows, and safety boundaries. This is your chance to apply design thinking in a real system where clarity, responsibility, and maintainability matter.
- Experiments: Help conduct experiments.
- Work interactively with LLMs: Use tools like GitHub Copilot and ChatGPT/GPT-4 as collaborative partners to (a) write and refine unit tests, (b) prototype decision modules and mission logic, (c) analyze design trade-offs and failure cases, (d) improve code readability, maintainability, and test coverage. However, you will take full responsibility for the final code — LLMs support your work, but critical thinking, review, and engineering judgment will be in your hands.
- Engineer for human-aware autonomy: While our focus is on drone autonomy, you will need to consider the role of the human-on-the-loop. Alongside developing autonomy capabilities, you will therefore be responsible for integrating features for (a) logging and explaining autonomous decisions, (b) designing interfaces for human-in-the-loop interaction, and (c) implementing safeguards for ambiguous or risky situations.
- Simulate, test, and iterate in our immersive Video Wall
While your code may later be deployed on physical UAVs, all acceptance tests will take place in our high-fidelity simulation environment, powered by our new Video Wall, which allows for immersive, real-time visualization of multi-drone missions. - Communicate your delivered solution orally and in writing.
Skills needed: You need to be reasonably proficient in Python and be interested in broader questions about AI-safety and autonomy on Cyber-Physical Systems.
Time commitment: This is a 3-credit hour research option. You cannot participate in this project for fewer credit hours. Further, there will be onboarding materials that all students will be expected to complete in the first 3 weeks of the semester, so that the remainder of the semester can be spent engaging in the selected/assigned projects.
Deliverables: These projects are most beneficial for students who have completed Paradigms and the AI-gateway course and want hands-on practice in building a larger AI-based system.
If you are interested in joining our team in the Fall, you can apply through this system and/or make an appointments to meet with me. You are far more likely to be offered a position if you come and talk to me (Prof. Jane Cleland-Huang).