In response to the technological gap in achieving full-process autonomous “pre-deployment-activation-flight” operations for traditional tube-launched multirotor systems within complex cross-media environments, this paper presents the design and implementation of a novel pre-embedded tube-launched quadrotor drone, designated “Feiying”. The primary objective is to overcome the limitations of conventional systems, which often suffer from response latency and fragmented stage control, thereby meeting the stringent demands for long-duration standby, rapid activation, and immediate response in challenging operational scenarios. The proposed solution incorporates several key innovations: a lightweight yet robust carbon-fiber and aluminum alloy composite load-bearing structure, a torsion-spring-driven folding arm mechanism for compact stowage, a hierarchical power management system for extended pre-embedded life, and a comprehensive state-machine-based control model governing the entire operational lifecycle. Experimental validation confirms that the Feiying quadrotor drone successfully achieves reliable pre-embedded standby for up to 30 days, coupled with a rapid activation and transition to autonomous flight within 0.1 seconds post-launch. This integrated design and control methodology provides a validated technical foundation for high-reliability deployment and high-speed response capabilities in pre-embedded tube-launched quadrotor drone applications.
The operational workflow of the Feiying quadrotor drone is initiated with its pre-embedment within a dedicated launch platform, where it remains in an ultra-low-power standby state. Upon receiving an activation command, the platform’s launch mechanism propels the drone into the air. Following ejection, the quadrotor drone autonomously unlocks its propulsion system and stabilizes its flight, proceeding to execute pre-programmed missions. This process imposes specific design requirements: 1) The quadrotor drone must feature a foldable, compact form factor to fit within the cylindrical launch tube. 2) It must sustain long-term pre-embedded operation with minimal power drain and activate instantaneously upon command. 3) Post-launch, it must achieve autonomous stabilization and flight without external intervention.
The overall architecture of the quadrotor drone is strategically partitioned into three primary sections: the Energy Bay, the Flight Propulsion System, and the Control Bay. This layout ensures balanced mass distribution along the vertical axis, which is critical for maintaining a stable attitude during the high-acceleration launch phase. The core load-bearing components are fabricated from carbon fiber composites and machined aluminum alloy to withstand significant launch-induced stresses. The Energy Bay houses the independent power sources, including the main flight battery and the separate pre-launch control system battery. The Flight Propulsion System, positioned centrally, employs an X-configuration quadrotor layout. Its defining feature is the foldable arm mechanism, which allows the arms to retract fully, transforming the entire assembly into a cylindrical profile for tube storage. The Control Bay, located at the base, contains the avionics responsible for orchestrating the complete operational sequence.
The foldable arm mechanism is pivotal for the stowability of the quadrotor drone. It comprises a rotation shaft clamp plate, torsion springs, spring limit posts, a central shaft, and arm connectors. The clamp plate serves as the structural anchor, mounting the mechanism to the drone’s central frame. Each arm is connected via a shaft and is biased by a torsion spring. During stowage inside the launch tube, the arms are folded, compressing the torsion springs and storing elastic potential energy. Upon ejection, this energy is released, driving the rapid and reliable deployment of the arms to their flight position. Mechanical hard-stops on the clamp plate precisely limit the rotation, ensuring the arms lock into the correct operational geometry for the quadrotor drone.
The electronic system of the quadrotor drone is deliberately segmented into two independently powered subsystems: the Pre-Launch Control System (PLCS) and the Flight Control System (FCS). This separation enhances reliability and facilitates targeted power management. The PLCS core is a microcontroller equipped with a wireless receiver module to listen for activation commands. It manages the FCS power state via a PWM-driven relay and establishes a USART communication link with the flight controller for command and status synchronization. It is powered by a dedicated 2S Li-ion battery, optimized for long-term, low-power standby.
The FCS is powered by a high-capacity 4S LiPo battery. Power flows through the PLCS-controlled relay, passes through a current sensor for monitoring, and is then conditioned by voltage regulators to provide 16.8 V for the propulsion motors and 5 V for the flight controller and sensors. The avionics suite includes a GNSS module with barometric compensation for positioning, a radio receiver for potential manual override or telemetry downlink, and the core flight controller. The interconnectivity of these systems is summarized in the hardware framework. Key design parameters for the quadrotor drone are listed in Table 1.
| Parameter | Symbol | Value |
|---|---|---|
| Total Mass | $m$ | 1.32 kg |
| Wheelbase (Diagonal) | $l$ | 350 mm |
| Pre-embedded Endurance | $t_{pre}$ | 30 days |
| Propeller Diameter | $D_p$ | 7 inch |
| Motor KV Constant | $K_V$ | 1700 rpm/V |
| In-flight Endurance | $T_{flight}$ | 8 min |
A critical challenge for a tube-launched quadrotor drone is the autonomous transition from a passive, high-speed projectile to a stable, hovering vehicle. Relying on remote commands or simple timers is unreliable due to communication latency and environmental variability. Therefore, an onboard, sensor-based autonomous arming and takeoff logic was developed. The strategy leverages the inertial measurement unit (IMU) of the flight controller. Post-activation but pre-launch, the FCS initializes and begins monitoring its state. The key trigger for motor arming is the velocity along the launch axis (z-axis in body frame). Based on empirical data from initial launch tests, a velocity threshold $v_{threshold} = 8.0 \, \text{m/s}$ was established. The decision model is:
$$ \text{Arm} = \begin{cases}
1, & \text{if } v_z > v_{threshold} \text{ and } t > t_0 \\
0, & \text{otherwise}
\end{cases} $$
Here, $t_0$ represents a brief time delay for launch platform clearance, preventing false triggers from tube ejection vibrations. This method ensures the quadrotor drone’s motors activate only after it has fully cleared the tube, its arms are deployed, and it has achieved sufficient upward velocity, marking a safe and stable point to begin active flight control.
The flight dynamics of the quadrotor drone are modeled using a standard six-degree-of-freedom rigid-body formulation. Defining the body-fixed frame $(x_B, y_B, z_B)$ and the inertial frame $(x_I, y_I, z_I)$, the equations of motion are:
$$ \begin{aligned}
\ddot{x} &= -T (\cos\psi \sin\theta \cos\phi + \sin\psi \sin\phi) / m \\
\ddot{y} &= -T (\sin\psi \sin\theta \cos\phi – \cos\psi \sin\phi) / m \\
\ddot{z} &= g – T (\cos\phi \cos\theta) / m \\
\ddot{\phi} &= \dot{\theta}\dot{\psi} \frac{I_y – I_z}{I_x} + \frac{\tau_x}{I_x} \\
\ddot{\theta} &= \dot{\phi}\dot{\psi} \frac{I_z – I_x}{I_y} + \frac{\tau_y}{I_y} \\
\ddot{\psi} &= \dot{\phi}\dot{\theta} \frac{I_x – I_y}{I_z} + \frac{\tau_z}{I_z}
\end{aligned} $$
where $T$ is the total thrust, $m$ is the mass, $(\phi, \theta, \psi)$ are the roll, pitch, and yaw Euler angles, $(\tau_x, \tau_y, \tau_z)$ are the body-frame moments, and $(I_x, I_y, I_z)$ are the moments of inertia. The control inputs for the quadrotor drone are the moments and thrust, which are generated by differentially varying the speeds of the four rotors. For an X-configuration quadrotor drone, the relationship between rotor speeds $(\omega_1, \omega_2, \omega_3, \omega_4)$ and the forces/moments is given by the mixer matrix:
$$ \begin{bmatrix} T \\ \tau_x \\ \tau_y \\ \tau_z \end{bmatrix} =
\begin{bmatrix}
c_T & c_T & c_T & c_T \\
-\frac{\sqrt{2}}{2}d c_T & \frac{\sqrt{2}}{2}d c_T & \frac{\sqrt{2}}{2}d c_T & -\frac{\sqrt{2}}{2}d c_T \\
\frac{\sqrt{2}}{2}d c_T & -\frac{\sqrt{2}}{2}d c_T & \frac{\sqrt{2}}{2}d c_T & -\frac{\sqrt{2}}{2}d c_T \\
c_M & -c_M & c_M & -c_M
\end{bmatrix}
\begin{bmatrix} \omega_1^2 \\ \omega_2^2 \\ \omega_3^2 \\ \omega_4^2 \end{bmatrix} $$
where $d$ is the arm length (distance from center to motor), $c_T$ is the thrust coefficient, and $c_M$ is the torque coefficient.
The control strategy is architected around a hierarchical state machine that manages the entire lifecycle of the quadrotor drone. The Pre-Launch Control state machine handles the low-power standby, activation, and system power-up sequences. It begins in a “Standby” state with the FCS powered off. Upon receiving a valid RF activation command, it transitions to “Activating,” closes the power relay to boot the FCS, and then sends a “Pre-Launch” command, moving to a “Ready for Launch” state where it awaits the physical launch trigger from the platform.
The Flight Control state machine takes over from the “Pre-Launch” state. It initializes sensors and waits in a “Pre-Launch Monitoring” state. Crucially, it continuously monitors the z-axis velocity $v_z$. Once the launch is detected and $v_z > v_{threshold}$, the system immediately triggers motor arming, exiting the forced “Lockdown” state. It then engages the attitude and position controllers to stabilize the quadrotor drone and execute its mission profile. The state machine incorporates robust error handling, with fallback paths to a safe “Standby” or “Fault” state for conditions like failed initialization, arming errors, or persistent unstable attitudes, ensuring the overall safety of the quadrotor drone operation.
To validate the design, a series of tests were conducted. The pre-embedment and activation sequence was successfully demonstrated in a lab environment. The system correctly transitioned from standby (all indicators off) to FCS powered-on (red LED on) upon receiving a wireless command, and finally to the pre-launch armed state (blue LED on), confirming the reliability of the command link and power management for the quadrotor drone.
Subsequent terrestrial launch tests provided critical data on the dynamic transition phase. Synchronized data logging captured the unlock status, acceleration, and velocity. The results showed that approximately 0.016 seconds after platform ignition, the quadrotor drone exited the tube. The arms deployed fully within 0.1 seconds, coinciding with the velocity exceeding the 8.0 m/s threshold. At this precise moment, the flight controller’s lockdown flag was cleared, activating the motors. Attitude data indicated transient angular displacements of less than 15 degrees during this transition, with the quadrotor drone achieving a stable hover within 3 seconds post-motor-start. Figure 11 illustrates the tight coupling between the velocity threshold crossing and the motor arming event.
A comparative analysis was performed against alternative arming strategies to quantify the benefits of the chosen velocity-threshold method. A high-fidelity simulation model of the quadrotor drone dynamics and control was developed, incorporating the mixer matrix, motor dynamics (modeled as a first-order system $A=-1/T_m, B=1, C=1/T_m, D=0$), PWM normalization, and cascaded PID controllers for attitude and position. Three baseline methods were simulated under identical launch conditions: (B1) Timer-based unlock, (B2) Single-threshold acceleration trigger, and (B3) Acceleration trigger with a time-hold window. The performance was evaluated against the experimentally derived benchmarks: arm deployment completion at ~0.1s, velocity threshold of 8.0 m/s, and attitude settling within 3s. The results, summarized conceptually, show that the velocity-threshold method (our approach) consistently provided the optimal alignment with the physical launch sequence, leading to the most reliable and timely activation of the quadrotor drone’s flight controls, minimizing transient attitude errors.
| Strategy | Description | Key Limitation | Performance vs. Our Method |
|---|---|---|---|
| B1: Timer-Based | Unlock after a fixed time delay. | Highly sensitive to launch variability; prone to unlocking too early (before clear) or too late. | Poor temporal alignment with physical launch sequence; often leads to larger attitude transients. |
| B2: Acceleration-Only | Unlock when |az| exceeds a threshold. | Vulnerable to false triggers from launch shocks/vibrations before full arm deployment. | Frequent false positives cause premature/unstable activation, disrupting the natural launch-arm-unlock sequence. |
| B3: Accel + Time Window | Accel trigger must be sustained for a time window. | Introduces additional lag, delaying necessary control response. | Increased time from launch to motor start, typically resulting in higher altitude loss and longer stabilization time. |
| Our Method: Velocity Threshold | Unlock when vz > v_threshold (after brief delay t0). | Requires accurate velocity estimation (provided by IMU integration). | Best alignment with “arm deployment complete & clear of tube” physical state; enables fastest, most stable transition to flight. |
In conclusion, the “Feiying” project successfully addresses the integrated design and control challenges for a pre-embedded tube-launched quadrotor drone. The synergistic combination of a robust composite-mechanical design, a hierarchical power and control architecture, and an intelligent state-machine-based autonomy stack enables reliable long-term pre-embedment, instantaneous activation, and a stable autonomous transition from launch to flight. Experimental validation confirms a 30-day standby capability and a sub-0.1-second response from launch to controlled flight. This work establishes a comprehensive technical foundation for deployable quadrotor drone systems requiring high reliability and rapid response in complex, cross-domain environments. Future work will focus on system miniaturization, enhanced swarm coordination protocols for multiple launched quadrotor drones, and testing in more realistic maritime conditions.