Air Lab | Carnegie Mellon University
Estimating Urban Wind Fields Using UAV-Based Measurements | 2020
Through real-world experiments we proved our method accurately estimates wind inlet conditions using the wind measurements from a flying UAV.
Goods Delivery Energy Productivity | 2019 - Present
- Creating a neural network to select motion primitives for a UAV to fly in windy urban environments
- Building and validating an energy model for autonomous unmanned ground vehicles
- Developing a path-planning algorithm which factors in risk, energy consumption, and wind
MAGICC Lab | Brigham Young University
UAV Gesture Commands | 2018-19
- Designed and trained a model to classify ten gestures with an accuracy of 95% using accelerometer and gyroscope measurements.
- Presented the research at the ICUAS 2019 conference
- Designed and tested intuitive gestures and behaviors for natural directing of a fleet of UAVs
- Submitted article to the Journal of Intelligent & Robotics Systems
Multi-Mission Project | 2017-19
- Developed a search algorithm for cooperating UAVs which maximizes area knowledge and the number of tracked targets using Gaussian process regressions
- Presented the research at the ICUAS 2018 conference
- Implemented a Gaussian Mixture Model Kalman filter for more accurate target tracking with heterogeneous sensors
AUVSI Student Unmanned Aerial Systems Competition
- Developed a robust RRT path planner for the AUVSI SUAS competition. This planner avoided obstacles while minimizing the waypoint capture error through ensuring long straight paths through waypoints.
- Also fabricated and repaired fixed-wing UAVs, created an image distortion correction program for letter and shape recognition, and many other tasks over the three years on the team.
- Implemented the autopilot from Small Unmanned Aircraft: Theory and Practice in Python. This includes controllers, estimators, a path planner, and a path manager.