Module 01 ended with a verified set of engineering requirements. Those requirements now drive the first major design decision: what shape is this vehicle going to be? Configuration selection is not a preference. It is a consequence — a direct output of the mission requirements, the payload, the operating environment, and the logistics constraints established in the CONOPS. The engineer who selects a configuration before reading the requirements is solving a different problem than the one they were given.
The four configurations and what drives each one
Every unmanned vehicle in production today is a variation of one of four basic configurations. Understanding the engineering rationale behind each — not just the marketing language — is essential to making a defensible configuration choice.
Fixed-wing conventional
A fixed-wing aircraft generates lift from forward motion interacting with a wing. It cannot hover, and it requires either a runway, a catapult, or a hand-launch and enough forward speed to become airborne. In exchange for that constraint, it offers the best endurance and range of any configuration at comparable power levels. The physics is straightforward: a wing moving through air at the right angle of attack generates lift more efficiently than rotating blades, which must continuously fight gravity to maintain altitude.
Wing loading — the ratio of aircraft weight to wing area — determines how fast the aircraft must fly to generate sufficient lift, and how sensitive it is to gusts. Lower wing loading means slower flight and gentler stall characteristics, but also more wing area to carry and more drag at higher speeds. The Group-3 UAV built for Tribe Aerospace ran a wing loading of approximately 14 kg/m² — low enough for a clean hand-launch by two people at modest speed, high enough to maintain control in 20-knot crosswinds during landing approach.
Multirotor
A multirotor generates lift and control entirely from variable-speed rotors — typically four, six, or eight. It can hover precisely over a point, take off and land vertically from any flat surface, and maneuver in tight spaces that would be impossible for a fixed-wing vehicle. Those capabilities come at a significant cost in efficiency: rotors must continuously work against gravity even when stationary, and the power required to hover scales poorly with vehicle weight.
For most multirotor designs, endurance is limited to 20–40 minutes at operational payload, regardless of battery technology advances, because the fundamental physics of hover are unforgiving. A multirotor carrying a 2 kg payload and weighing 6 kg total must generate 6 kg of continuous thrust just to stay airborne, and generating that thrust consumes far more energy per unit time than a fixed-wing aircraft cruising at the same total weight.
The correct use case for multirotor is operations that genuinely require the ability to hover: close-range building inspection, precision delivery, operations in confined urban areas, and applications where the sensor must look straight down and remain stationary over a point. If the mission can be accomplished without hovering, a fixed-wing vehicle will almost always be more capable and more efficient.
Hybrid VTOL fixed-wing
The hybrid VTOL attempts to combine the vertical takeoff and landing capability of a multirotor with the cruise efficiency of a fixed-wing aircraft. In the most common implementation — the tail-sitter and the tilt-rotor aside — the vehicle has a fixed wing for cruise and a separate set of vertical lift rotors that are only powered during takeoff, landing, and hover.
The engineering challenge of a hybrid VTOL is the weight and complexity penalty of carrying two propulsion systems. The vertical lift rotors are dead weight during cruise. The fixed wing is irrelevant during hover. The vehicle must be designed to be aerodynamically stable at all points of the transition from hover to forward flight — a regime that is neither cleanly hovering nor cleanly flying, and that generates loads on the airframe that neither pure configuration has to manage.
The X8 VTOL ISR platform designed for a European client demonstrates where the hybrid makes sense: a mission requiring runway-independent operations in areas without launch infrastructure, with a sensor that needs both hover capability (for target examination) and long-range transit (to reach the area of operations). The design used a gas engine for cruise endurance and electric motors for the VTOL phase, keeping the gas engine out of the mechanically demanding hover regime.
Flying wing and blended wing
The flying wing eliminates the fuselage and tail surfaces, generating both lift and pitch control from a single continuous wing surface. At the right scale and mission profile, it is the most aerodynamically efficient configuration available — the entire planform contributes to lift, and there is no fuselage drag. It is also inherently difficult to stabilize, since there is no conventional tail to provide pitch restoring force, and payload volume is constrained by the wing cross-section.
Flying wing configurations work best for high-altitude, long-endurance ISR where the sensor can be accommodated in a pod or in the center-section, where the mission profile is predominantly straight-line cruise, and where the operator has the avionics sophistication to manage an inherently less stable platform. The Group-2 AFRL jet swarm platform used a modified flying wing configuration specifically because the mission was long-range transit with a relatively small forward-looking sensor — exactly the profile the configuration is designed for.
Reynolds number effects at UAS scale
Aerodynamic data from full-scale aviation does not transfer cleanly to small unmanned systems. The Reynolds number — a dimensionless quantity representing the ratio of inertial to viscous forces in the airflow — governs how air behaves around a surface, and small UAS operate at Reynolds numbers between 50,000 and 500,000, compared to 10,000,000 or more for full-scale aircraft.
At low Reynolds numbers, the boundary layer — the thin region of air immediately adjacent to the wing surface — behaves differently. It separates more readily from the surface, particularly on the upper surface of a wing at moderate angles of attack, forming a laminar separation bubble that degrades lift and dramatically increases drag. Airfoils that perform efficiently at full scale can exhibit poor characteristics at UAS scale for this reason.
The practical consequence is that airfoil selection for a UAS must be based on low-Reynolds-number data, not standard aviation data. Airfoils developed specifically for this regime — such as those in the Eppler and Selig databases — exhibit flat-bottom or reflexed camber lines that maintain attached flow at low chord Reynolds numbers. CFD validation at the correct Reynolds number is not optional; it is the only way to predict actual performance before the first part is manufactured.
Wing design: aspect ratio and wing loading
Aspect ratio (AR) is the ratio of wing span to mean chord, or equivalently span² divided by wing area. High aspect ratio wings — long and narrow — produce less induced drag for a given amount of lift, improving cruise efficiency and endurance. Low aspect ratio wings are structurally stiffer for the same span, less sensitive to gusts, and easier to transport. The Group-3 UAV used an aspect ratio of approximately 9 — high enough for good endurance performance, low enough for manageable structural design in carbon fiber.
Wing loading (W/S, in kg/m² or N/m²) determines stall speed, gust sensitivity, and maneuverability. Lower wing loading produces lower stall speeds and more docile handling — ideal for hand-launch vehicles that must fly slowly enough for safe recovery. Higher wing loading produces higher cruise speeds and less gust sensitivity — appropriate for tactical systems that must maintain precision in turbulent low-altitude environments. A typical sport UAV runs 8–15 kg/m²; a tactical fixed-wing may run 25–40 kg/m².
Airfoil selection and CFD methodology
Airfoil selection begins with the operating Reynolds number range and the required lift coefficient at cruise. For a given wing loading and cruise speed, the required cruise lift coefficient CL is fixed: CL = 2W / (ρV²S). An airfoil that cannot efficiently generate that CL at the operating Reynolds number will require either a larger wing or higher speed, both of which have downstream consequences for the rest of the design.
The selection process in practice: identify three to five candidate airfoils from the low-Reynolds-number literature, run 2D panel code analysis (XFOIL is the standard tool) to compare their lift, drag, and pitching moment characteristics across the operating range, select the top two candidates, and run 3D CFD on the proposed wing geometry to capture tip effects, span loading distribution, and the behavior near stall. The CFD results set the final airfoil and wing geometry before any tooling is designed.
For the Group-1 loitering munition redesign, the original wing airfoil was replaced with a custom profile derived from XFOIL optimization at the operating Reynolds number, then validated in ANSYS Fluent at the full wing geometry. The redesigned wing produced 18% more lift at cruise angle of attack and delayed stall by approximately 3 degrees compared to the original — a meaningful improvement in both performance and safety margin, achieved without changing the wing planform dimensions.
Stability: static margin and CG management
Aerodynamic stability in a fixed-wing aircraft requires that the center of gravity (CG) be forward of the neutral point (NP) — the aerodynamic equivalent of the center of pressure for a complete aircraft including the tail. The distance between CG and NP, expressed as a fraction of the mean aerodynamic chord, is the static margin. A positive static margin means the aircraft will naturally return to trimmed flight after a disturbance; a negative margin means it will diverge.
Static margin for a hand-launched UAV is typically set between 8% and 15% MAC. Too little margin produces an aircraft that is twitchy and demands a high-bandwidth autopilot to maintain control. Too much margin produces an aircraft that is sluggish, requires large control deflections to maneuver, and wastes energy fighting its own stability. For an autopilot-controlled vehicle with modern flight control software, static margins toward the lower end of this range are preferable — the autopilot can handle the faster dynamics, and the reduced stability margin allows more efficient flight.
CG management is a continuous design task, not a one-time calculation. As the payload changes, as propellant is consumed, as batteries discharge, and as the vehicle configuration changes between missions, the CG moves. The design must accommodate the full range of expected CG positions while maintaining acceptable static margin throughout. This is why the avionics bay location, battery position, and payload bay position are not independent design choices — they are a coupled system that must be solved simultaneously.
Control surfaces and sizing
Control surfaces — ailerons, elevators, rudders, and elevons on flying wings — must be large enough to generate the moments required for the demanded maneuvers, while being small enough not to create excessive drag when deflected. The standard initial sizing rule is that control surfaces should occupy 25–35% of the chord over the span they cover, with maximum deflection angles of ±20–25 degrees before aerodynamic effectiveness begins to decrease due to flow separation.
For a UAV with an autopilot, the minimum control authority requirement is typically set by the crosswind landing or crosswind cruise condition specified in the requirements. If the vehicle must maintain controlled flight in a 20-knot crosswind, the control surfaces must generate sufficient roll and yaw moment to counteract the imposed side force and moment at maximum crosswind. This is verified in CFD at the maximum deflection angle and maximum crosswind condition before the design is released for manufacturing.