CSE - Drone Response: Engineering multi-UAV Autonomous Systems
The DroneResponse Lab focuses on developing advanced multi-vehicle autonomous UAV systems to support critical emergency response missions, such as search and rescue, disaster assessment, and recovery operations. Our research integrates cutting-edge technologies including computer vision, real-time anomaly detection, autonomous decision-making, and situational awareness for humans. For a quick overview of our Drone Response project please see: https://youtu.be/DyKqxkesgg0?si=OnNHl0Y6dSzn_Ap-
We have opportunities for several different projects for undergraduate students - some focusing more on development and others more on experimental research. Most of our projects (except the HCI one listed below) include a significant amount of Python programming (at varying levels of difficulty), therefore you must have proficient Python programming skills to apply for these positions. You may also apply for a project with a partner if you wish to (both of you must name the other person).
- Computer Vision: You will be helping to train new aerial Computer Visions models for use by the drones. The project will include some annotation (10%), programming (60%), running experiments (15%), writing up results (10%), and meetings (5%). [Credit or paid position is possible. Paid position would focus more on annotation and experimentation, with some programming. Paid position can be 5-8 hours to week and no experience is needed]
- Enhancing our LLM-based Natural Language Interface: You will be enhancing our NL interface for supporting emergency responders in providing mission inputs based on voice and text. The inputs need to be mapped to system capabilities. The project will include some programming (70%), running experiments (15%), writing up results (10%), and meetings (5%). [Credit only]
- RAG (Retrieval Augmented Generation): You will be doing research in how to dynamically retrieve and analyze documents describing best-practices for diverse emergency response scenarios. You will also use LangChain to coordinate these tasks, and you will experiment with ways to map best practices into a structured format. This project is more research oriented, so you should be willing to think of ideas, experiment with them, and report your ideas. The project will include reading research papers (20%), programming (50%), running experiments (15%), writing up results (10%), and meetings (5%). [Credit only]
- Human Computer Interaction: Situational Awareness: This project will involve organizing and conducting user studies focused around situational awareness in multi-vehicle drone systems. The studies will all be conducted in our simulation environment, with possible outdoor field tests in the Spring. The project will include reading research papers (10%), programming (20%), running experiments (50%), writing up results (15%), and meetings (5%). [Credit or paid position is possible]
- Mobile Apps: We have the need to develop several tablet-based and mobile apps for supporting human operators of drones. Please ONLY apply for this project if you are proficient at mobile app development, and looking for the opportunity to engage in a realistic real-world project. The project will include eliciting requirements (10%), design (15%), programming (60%), testing (10%), and meetings (5%). [Credit or paid position is possible]
If you are interested in any of these positions, 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).
The DroneResponse Lab is a dynamic and exciting place to work because it sits at the intersection of cutting-edge technology and real-world impact. Our research focuses on developing autonomous UAV systems that can support emergency response efforts, which is both technically challenging and societally crucial. The lab works on integrating autonomous decision-making, multi-vehicle coordination, real-time data analysis, and regulatory compliance to create systems that can operate effectively in complex, high-stakes environments like disaster zones and search-and-rescue missions