Quad-Rotors and Control Experiments Lab

Research Question

How can undergraduate students bridge the gap between theoretical control design and real-world experimental validation, using quad-rotor platforms as a testbed for state estimation, control, and sensor integration?

Method

This project was an undergraduate teaching laboratory at ETH Zurich’s Institute for Dynamic Systems and Control (IDSC), where I served as lecturer and laboratory lead. The course combined:

• Control theory lectures — covering quad-rotor dynamics, state estimation (Kalman filtering), control design (PID, LQR, MPC), and sensor integration (IMU, optical flow, GPS).
• Hands-on quad-rotor experiments — students implemented state estimation and control algorithms on physical quad-rotor platforms, working with real sensor noise, actuator limits, and flight dynamics.
• Experimental validation methodology — students learned to design experiments, collect data, diagnose control performance, and iterate on their designs based on empirical results.
• Safety and operational procedures — laboratory protocols for safe quad-rotor operation, emergency procedures, and experimental documentation.

The course served approximately 120 undergraduate students per year, making it one of the larger hands-on control laboratory courses at ETH Zurich.

Personal Contribution

Frederik Banis was the principal teacher and lab manager for the Quad-Rotors and Control Experiments Lab (2023–2024). He was responsible for lecture design, laboratory setup and maintenance, supervision of student groups, assessment, and ensuring safe laboratory operations. He also coordinated the parallel Control Experiments Lab at IfA/PSL (~200 students), managing experimental setups across multiple course sections.

Validation and Key Results

• The course successfully enabled ~120 students per year to implement and test control algorithms on physical hardware, achieving the pedagogical goal of bridging theory and practice.
• Students produced functioning quad-rotor control implementations, demonstrating state estimation, trajectory tracking, and disturbance rejection on real platforms.
• The laboratory served as a proof-of-concept for integrating research-grade experimental platforms into undergraduate control education.

Outputs

• Teaching: Undergraduate course at ETH Zurich, 2023–2024.
• Related infrastructure: Laboratory setup for quad-rotor control experiments, including hardware platforms, sensor suites, and safety protocols.

Collaborators

• ETH Zurich, Institute for Dynamic Systems and Control (IDSC) — course hosting and laboratory infrastructure.
• ETH Zurich, IfA/PSL — parallel Control Experiments Lab coordination.

Status and Next Steps

Status: Historical (2023–2024). The course continues at ETH Zurich; my role concluded with the end of my appointment.

Transfer to current research: The quad-rotor laboratory demonstrates the closed-loop experimentation methodology that is central to the research program. The same principles — sense, estimate, decide, act, validate — that students applied to quad-rotor control now underpin adaptive biomedical sensing and closed-loop discovery. The laboratory experience also informs the teaching philosophy of integrating theory, computation, and physical experimentation.

See the Teaching page for the full teaching record and the Methods page for the closed-loop experimentation methodology.