As educators in aeronautical engineering, we have long recognised the transformative potential of unmanned aerial systems in both military and civilian domains. Inspired by the directive to strengthen unmanned combat research and accelerate the training of drone operators and commanders, our institution has placed a heightened emphasis on the development of comprehensive practice courses that accompany theoretical instruction. This article details our experience designing and implementing a comprehensive practice course centered on fixed-wing drone, with the goal of integrating knowledge, skills, and professional qualities through hands-on project-based learning.
1. Introduction
The aviation pioneer Otto Lilienthal once remarked, “To invent an airplane is nothing. To build one is something. But to fly it is everything.” This sentiment encapsulates the essence of our pedagogical philosophy: the ultimate validation of any aircraft design is its ability to fly. Fixed-wing drone, in particular, demands a rigorous synthesis of aerodynamics, flight mechanics, control theory, communication systems, and structural integration. Unlike multirotor platforms, fixed-wing drone present greater challenges in aerodynamic calculation, flight control, and telemetry, thereby inspiring students to engage with deeper technical complexities. Our comprehensive practice course was therefore conceived around a single, compelling objective: to get a fixed-wing drone airborne, and to use that process as a vehicle for systematic learning.
In this paper, we present the design, implementation, and assessment of this practice course, drawing on our experience over multiple semesters. We discuss the basic philosophy, content structure, instructional techniques, evaluation methods, and lessons learned. Throughout, we emphasise the repeated use of the term ‘fixed-wing drone’ to underscore its centrality to the curriculum.
2. Design of Comprehensive Practice Teaching
2.1 Basic Philosophy
Our core principle is that “flying the fixed-wing drone” serves as the unifying thread for all sub-practices. The fixed-wing drone we use is an electrically powered model with a wingspan of 1.8 m and a mass of 2 kg. Its geometric parameters are known, but its aerodynamic characteristics must be calculated by students. The practice covers the entire lifecycle of a drone project: conceptual design, aerodynamic analysis, flight control simulation, wireless communication testing, final assembly, and flight testing. This approach not only reinforces theoretical knowledge but also fosters teamwork, engineering judgment, and a sense of responsibility. The choice of fixed-wing drone over multirotor is deliberate: its broader application in military reconnaissance, surveillance, and precision agriculture demands a deeper understanding of aerodynamics and stability, thereby stretching the capabilities of our students.
2.2 Content Arrangement
The comprehensive practice comprises five main modules, as summarised in the table below. Each module is designed to be completed over several sessions, with clear deliverables and checkpoints.
| Module | Description | Key Activities | Duration (hours) |
|---|---|---|---|
| 1. Aerodynamic Analysis | Computing lift, drag, moment coefficients using AVL (Athena Vortex Lattice) software | Build 3D geometry; run vortex lattice method; extract stability derivatives; determine trim condition | 8 |
| 2. Flight Control Simulation | Constructing a linear small-perturbation model and designing control laws in Simulink | Derive state-space equations; tune PID gains; simulate closed-loop response; evaluate handling qualities | 10 |
| 3. Communication System Testing | Setting up and characterising a line-of-sight radio link | Configure frequency, baud rate, and power; measure latency and range; test interference scenarios | 6 |
| 4. Full-vehicle Assembly and Integration | Integrating airframe, avionics, and autopilot hardware | Assemble wing, fuselage, tail; install servos, receiver, flight controller; perform pre-flight checks | 8 |
| 5. Flight Test and Data Analysis | Conducting manual and autonomous flights, collecting telemetry | Pre-flight checklist; remote control flight; switch to autonomous mode; analyse logged data | 6 |
In the aerodynamic module, students build a geometric model of the fixed-wing drone and compute its lift curve slope. For example, the lift coefficient can be expressed as:
$$C_L = C_{L0} + C_{L\alpha} \alpha$$
where $\alpha$ is the angle of attack. Using AVL, they obtain the pitching moment coefficient $C_{m}$ and locate the aerodynamic centre. A typical result for our fixed-wing drone is:
$$C_{m\alpha} = \frac{dC_m}{d\alpha} \approx -0.12 \text{ per degree}$$
which confirms static longitudinal stability. Students then calculate the required flight speed for level flight:
$$V = \sqrt{\frac{2W}{\rho S C_L}}$$
where $W$ is weight, $\rho$ air density, and $S$ wing area.
The flight control module introduces the small-perturbation linear model. The longitudinal dynamics are described by:
$$\begin{bmatrix} \dot{u} \\ \dot{w} \\ \dot{q} \\ \dot{\theta} \end{bmatrix} = \begin{bmatrix} X_u & X_w & 0 & -g\cos\theta_0 \\ Z_u & Z_w & u_0 & 0 \\ M_u & M_w & M_q & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix} \begin{bmatrix} u \\ w \\ q \\ \theta \end{bmatrix} + \begin{bmatrix} X_{\delta_e} \\ Z_{\delta_e} \\ M_{\delta_e} \\ 0 \end{bmatrix} \delta_e$$
Students tune proportional-integral-derivative (PID) gains for pitch attitude hold and altitude hold. They simulate step responses and observe how changes in gains affect settling time and overshoot.
For the communication module, students use a 433 MHz telemetry radio module. They measure received signal strength indicator (RSSI) as a function of distance in an open field. The path loss model is approximated by:
$$P_r(d) = P_t G_t G_r \left(\frac{\lambda}{4\pi d}\right)^2$$
They also perform jamming tests: one group attempts to disrupt another group’s link, forcing the defending group to adapt by switching frequencies or adjusting power.
The assembly module requires meticulous attention. Students follow a structured pre-flight checklist that includes verifying control surface deflections, neutral trim, battery voltage, and autopilot initialisation. They document each step with photographs and tables.
Finally, the flight test module is the climax. Each team attempts a manual takeoff, a stabilised loiter, and a fully autonomous waypoint mission. Post-flight, they download flight logs and analyse altitude hold accuracy, heading error, and control surface activity.
2.3 Design Techniques
Several instructional strategies have proven effective in maximising student engagement and learning outcomes.
Group-based implementation. Teams of four to five students are formed, with roles allocated for aerodynamicist, control engineer, communication specialist, and test pilot. Peer evaluation is incorporated to discourage free-riding. The fixed-wing drone project inherently demands collaboration, as errors in one module (e.g., wrong aerodynamic data) cascade into later stages.
Repeated drills and reinforcement. Pre-flight checks and system boot-up procedures are rehearsed multiple times. We require each student to perform the checklist on three separate days before the actual flight. This repetition ingrains safety habits and reduces the likelihood of mistakes. For example, students verify that the elevator servo moves in the correct direction—a critical check that, if omitted, could lead to a crash.
Guiding and “pit-digging”. Instructors provide initial guidance to set the direction, but also intentionally plant subtle errors (“pits”) for students to discover. For instance, in the control simulation module, we assign different initial PID parameters to each team; some are deliberately unstable. Students must debug and correct them. In the communication module, we secretly swap one team’s antenna with a damaged one, forcing them to diagnose the poor range. The “pit” size is calibrated: too small and students overlook it; too large and they become discouraged. Over successive offerings, we have refined these obstacles based on observed student responses.
Competitive adversarial scenarios. To inject realism and excitement, we introduce friendly competition. During the communication module, one team acts as the “red force” and attempts to jam or spoof the link of the “blue force”. The blue team must counteract by adjusting frequency channels or employing spread spectrum techniques. This adversarial element boosts motivation and provides a taste of electronic warfare. Scores are adjusted accordingly—teams that successfully defend their link earn bonus points.
3. Assessment of Comprehensive Practice
3.1 Project Report Design
Each module culminates in a written deliverable that integrates data, analysis, and reflection. The project report template includes four major sections:
- Aerodynamic Analysis Report: Includes AVL geometry screenshots, computed lift and moment coefficients, stability margin, and a comparison with theoretical estimates.
- Control Simulation Report: Presents the Simulink model, tuned PID gains, step response plots, and discussion of how gains affect stability.
- Communication System Report: Documents radio configuration parameters, RSSI vs. distance data, latency measurements, and any interference encountered.
- Assembly and Flight Report: Contains photos of each assembly step, pre-flight checklists, flight logs, and post-flight data analysis (e.g., altitude deviation during autonomous flight).
Students are required to label all figures and tables, and to include a section on lessons learned. The report contributes 40% of the final grade.
3.2 Analysis of Assessment Results
We have collected assessment data from two consecutive cohorts (total 48 students). Each of the four sub-practices was scored out of 20 points. The average scores and distribution are shown in Table 2 and Figure 1 (conceptual distribution).
| Module | Mean | Maximum | Minimum | Standard Deviation |
|---|---|---|---|---|
| Aerodynamic Analysis | 18.0 | 20 | 16 | 1.2 |
| Control Simulation | 16.1 | 20 | 12 | 2.5 |
| Communication System | 14.8 | 17 | 12 | 1.8 |
| Assembly and Flight | 19.05 | 20 | 18 | 0.9 |
Several observations emerge. The aerodynamic analysis and assembly/flight modules yield high averages with narrow variance. This is partly because these tasks are well-defined and steps are clearly prescribed; students largely succeed if they follow instructions carefully. In contrast, the control simulation and communication modules exhibit lower averages and wider spreads. The “pit-digging” component accounts for this: some teams fail to identify and correct the intentional errors, resulting in flawed simulations or failed communication tests. Notably, the team that underwent adversarial jamming in the communication module scored an average of 16.0, higher than the global average of 14.8. This suggests that the competitive element sharpens focus and forces deeper engagement.
The score distribution for the communication module (conceptual) is shown below. The histogram is not strictly normal, indicating that the challenge level may not be perfectly calibrated.
Figure 1. Score distribution for the communication module. The x-axis represents score ranges (12–14, 14–16, 16–18, 18–20) and the y-axis the number of students.
4. Reflections and Considerations
4.1 Weight of Practical vs. Test Scores
Traditional theoretical courses rely heavily on test-based evaluation. In contrast, our comprehensive practice course originally assigned 80% of the grade to practical execution (following step-by-step instructions) and only 20% to a final written test. This led to grade inflation and insufficient differentiation. For instance, in the aerodynamic and flight modules, nearly all students scored above 18/20, masking differences in understanding. We are therefore adjusting the weighting: increasing the test component to 40% by incorporating short in-lab quizzes on operational nuances and troubleshooting scenarios. These quizzes include questions like, “What is the effect of increasing the proportional gain on elevator response?” and “If the telemetry link shows high latency, which parameter would you check first?” This balances practical skills with conceptual comprehension.
4.2 Trade-off between Obstacle Setting and Practical Risk
The “pit-digging” technique enhances learning but must be carefully managed to avoid compromising flight safety. In the aerodynamic and flight modules, any accidental error could lead to an actual crash. Therefore, we limit obstacles in these modules to low-risk stages. For example, in the assembly module, we might intentionally miswire a servo connector so that the control surface moves backwards. The student discovers this during the pre-flight check and corrects it—no harm done. However, we never alter parameters that could cause structural failure or loss of control during flight. In the communication module, jamming is conducted at low power and only after the defending team has already confirmed a stable link. We also maintain a safety supervisor (a teaching assistant) who monitors all operations and intervenes if a team appears about to make a dangerous mistake.
Over time, we have developed a catalogue of “safe pits” that are documented and reused across semesters. Each pit is rated for difficulty and risk, and only those with risk level ≤ 2 (on a 1–5 scale) are permitted. This ensures that the learning benefit outweighs the potential for mishaps.
4.3 Further Improvements
Future iterations of the course will incorporate more advanced topics. For instance, we plan to introduce multi-sensor fusion (GPS+IMU+barometer) and robust control techniques for the fixed-wing drone. We also aim to increase the autonomy level: instead of simple waypoint navigation, students could design a loiter pattern with wind compensation. Another direction is to integrate a small payload (e.g., a camera) and conduct a simulated reconnaissance mission, thereby exposing students to real-world operational constraints.
Additionally, we are developing a virtual reality (VR) simulation environment for pre-flight rehearsal. This would allow students to practice takeoff and landing without risking the actual fixed-wing drone. The simulator could also be used to introduce abnormal scenarios (e.g., engine failure, GPS denial) that are too dangerous to try in reality.
5. Conclusion
The comprehensive practice course centered on fixed-wing drone has proven to be a powerful vehicle for bridging theory and practice. By organising the curriculum around the compelling goal of flight, we engage students deeply across multiple disciplines. The combination of aerodynamic analysis, control simulation, communication testing, assembly, and real flight creates a rich learning ecosystem. Our use of group work, repeated drills, intentional pitfalls, and adversarial challenges fosters teamwork, critical thinking, and resilience. Assessment data reveal that modules with higher cognitive demands and “pit-digging” yield more differentiated scores, which is desirable for fair grading. At the same time, we must carefully balance the risks inherent in hands-on practice. As we continue to refine the course, we remain committed to the principle that learning to design, build, and fly a fixed-wing drone is an unparalleled educational experience that prepares students for the future of unmanned aviation.