Multi-Agent Robotics Testbed
Multi-Agent Robotics Testbed (Crazyflie Platform)
A testbed that lets several micro-drones fly together indoors, built to study how a group of robots can learn to coordinate on its own.
- Role
- Visiting Student Researcher
- Period
- Jan. 2026 – Apr. 2026
- Institution
- North Carolina State University Supervised by Dr. Yuchen Liu
Why this matters
Most multi-agent reinforcement learning is validated only in simulation. What works in simulation often fails on real hardware — localization drifts, radio links lag, batteries sag. Testing whether an algorithm actually works requires a physical platform that can reproduce experiments reliably.
What I did
- Built and maintained a Crazyflie-based multi-drone testbed with indoor localization using the Loco positioning system.
- Developed Python-based control and logging pipelines using crazyflie-lib-python for real-time multi-drone coordination and data collection.
- Conducted experiments on multi-agent reinforcement learning and swarm control in both simulation and physical drone environments.
Technologies
- Python
- crazyflie-lib-python
- Loco Positioning
- Multi-Agent RL
- Swarm Control