DECEMBER 2023 - MARCH 2025
Full research project title:
Taming the Oblique Wing: Improving Fuel Efficiency by Developing and Flight Testing an Oblique Wing Aircraft Utilizing a Novel Control Method
I designed, optimized, built, flew, and analyzed an oblique wing model airplane to mitigate the flight control problems of oblique wings, a previously unresolved issue preventing the implementation of this highly efficient design.
This was an independent project I conducted at home. The purpose was both to advance both the aerospace field and for myself to learn more about the engineering design process.
This project won a total of over $100,000 in awards through various competitions, including the International Science and Engineering Fair (ISEF), the Regeneron Science Talent Search, and the Davidson Fellows program. The project has been featured in Forbes, by ARC Seattle, and by the Society of Professional Engineering Employees in Aerospace (SPEEA). I also was invited to present this project at the 2024 Pacific Northwest AIAA Technical Symposium and at the 2025 AIAA Region VI Conference.
Design Techniques: Computational Fluid Dynamics (CFD), Computer Aided Design (CAD), Printed Circuit Board (PCB) Design
Manufacturing: 3D Printing, Laser Cutting, Soldering, Hand-Assembly
Software Applications: Fusion 360, Autodesk CFD, EasyEDA, Inkscape, Arduino IDE, Visual Studio Code, Git/Github
Programming Languages & Tools: C++, Python, Matplotlib, dRehmFlight
This powerpoint was used to present my project at an AIAA conference. It provides a high level summary of the entire project, details its significance, and is useful for contextualizing the various engineering components used in this project.
Oblique Wing Slides Edited.pdfPictures and descriptions from throughout the project.
2D CFD simulation of the SC(2)-0610 airfoil at 4° angle of attack at various airspeeds to find the critical Mach number.
Full 3D CFD mesh (approx. 5M elements), used to approximate the efficiency gains of a full sized oblique wing airplane.
Identifying locations of supersonic flow over the oblique wing and adjusting in the next iteration. Regions of supersonic flow to mitigate are shown in orange.
Static pressure plots comparing the conventional, symmetric wing design (left) against the oblique wing design (right). Conditions set to Mach 0.82 at 10,000m altitude. Sizing based on the A320 aircraft.
Results of the CFD analysis of the full-scale aircraft.
CFD convergence plot, used to verify that the simulation has reached steady state.
2D CAD drawings of an early prototype of the airframe.
The CAD model of the final airframe (third prototype). 3D printed parts were used for complex curves or load bearing areas, and foam board was selected for aerodynamic surfaces for its low weight and durability.
Picture of me with the final design.
Top down view of wing pivot mechanism and wing spar in CAD.
CAD model of the wing pivot mechanism. The pivot actuates the wing between 0 and 30 degrees of wing sweep while supporting the lift forces of the oblique wing.
Side view of the the wing pivot mechanism in CAD. The entire mechanism rotates with only one moving part for simplicity.
PCB schematic, showing the main connections to sensors, actuators, and other I/O pins.
Flight computer PCB final design. Above (in red) is the top layer, below (in blue) is the bottom layer.
The flight computer installed in the aircraft.
Above is a block diagram of the control loop implemented on the flight computer. To the right expands more detail on the custom anti-coupling flight control algorithm developed through this project.
Building the first prototype.
Soldering the flight computer.
Hand-launching a prototype.
First prototype:
Successfully flew under self-power
Demonstrated manually controlled flight
Demonstrated wing pivoting on the ground, but was unable to sustain flight with the wing at an oblique angle on the first few flights
Second prototype:
Tuned wing to pivot to exactly 30 degrees, based on CFD results
Integrated custom flight computer and collected data to tune the new flight control algorithm
Increased wing spar strength and reduced dihedral angle following wing-folding induced crashes
Reduced wing pivot speed to minimize creating asymmetric lift as the wing adjusts its oblique angle
Removed protective motor shroud, as it was limiting the thrust of the ducted fans
Added paint to protect it from the weather and wet grass
Demonstrated flight-computer-augmented pilot control, sustained oblique wing flight, and improved control of oblique wing aircraft
Third and final prototype:
Redesigned wing pivot mechanism
Improved wing shape with smoother airfoil construction and rounded wingtips
Simplified design down to one hatch instead of two
Removed crash-preventing landing gear
Rebuilt from the ground up to remove deformations and defects from previous crashes and last-minute modifications
Aftermath of a crash. This crash was due to aeroelastic instability of the forward wing, causing it to fold. This was resolved by stiffening the wing spar.
Aftermath of another crash. This crash was due to a low-speed stall, partly due to pilot error. After rebuilding, I learned to approach my landings at a higher airspeed.
Another crash due to aerodynamic coupling induced oscillations while testing the oblique wing without the flight computer. Notice how as the airplane pitches, it rolls in the opposite direction, showing coupling behavior. With the flight computer implemented, these oscillations were mitigated, as seen in the data below.
Sweep testing on the ground. Photos of the wing at 0 degree sweep and 30 degree sweep superimposed.
Initial mid-air wing sweep testing. The two photos were taken at two moments in the flight and stitched together to demonstrate the wing sweep change. Lighting and quality are not great on this photo, as it was taken from far away with a smartphone.
Sustained oblique wing flight with the custom flight controller stabilizing the airplane.
Another view of oblique wing flight.
Scatterplot of roll rate versus roll command as an example of the data collected. Notice how the regression line for data from flights with the flight computer has not only a higher R^2, but also has a higher slope, further indicating increased controllability.
Improved control authority demonstrated by comparing R2 values of aircraft response rate versus pilot command, with and without the flight computer
Diagram of coupling behavior without flight computer (control). Notice the strong coupling between roll and pitch.
Diagram of coupling behav ior with flight computer (experimental). Notice the significant reduction in coupling between roll and pitch. Roll-yaw intentionally remains coupled for coordinated turning.
Q-Q Plots for all attitude flight data collected. The data points closely follow a straight line, showing normality and validating statistical tests
Shen_Kevin_Paper.pdf
