Data-Driven Control for Agile Flight in a Confined Space
C++, Python, ROS2/ROS, Aerial Robots, Motion Planning, Computer Vision, Dynamic Systems
Overview
Personal Motivation
MetaFly by Bionic Bird
How hard is it to fly this thing?
System Design
Transmitter - Feedforward
Motion Capture - Feedback
The Closed Loop
Skip to Control Architecture which discusses the various control strategies that work.
Experiments
Screw Trajectory
Data Collection
Control Architecture
Feedback Altitude Control
Feedback Roll Control
Feedback Turning Radius Control
Feedback Yaw Control
Motion Planner
Negative Results and Discussion
Acknowledgements
Thanks Jake, for mentoring me throughout this project, and for being so nice. You’re someone I can really talk to :)
Thanks Davin, for helping me build a totally epic DRONE CAGE!!
Thanks Drew, for helping me set up and debug the OptiTrack system.
This Post is under construction 🛠️
Building a perfect portfolio from the ground up is a demanding process.
Over the coming weeks, I will be adding more information about my personal and professional projects.
If you’re interested, please check back on July 30th, 2024.
Your curiosity is greatly appreciated :)
In the meanwhile, the posts on:
- Search-and-Rescue Robot Dog: Autonomous Exploration and Visual SLAM with Unitree Go1,
- Complete Extended-Kalman-Filter SLAM Pipeline in C++ from Scratch,
- Dexterous Manipulation through Virtual Reality,
- Optimized Torques for Load Carrying,
- Dancing Hexapod - Full SE3 Body Pose Control,
- Navigation and Kalman Filter Localization of Self Driving Car,
- Mobile/Whole-body Manipulation with KUKA youBot,
- Legged Robots for Object Manipulation: A Review,
- TeleOperation of Robot Swarms,
- Home Automation with Clap Sequence Control,
- Robot playing T-Rex Game, and
- Maze Solver
are complete and ready for viewing!