Aerodynamics of small flyers

Dandelion seeds can travel more than 150 km and stay airborne for days without any energy source. Our group discovered how: the bristly pappus generates a separated vortex ring, a new class of fluid flow that we reported in Nature in 2018. This PhD will exploit this mechanism to design the next generation of flying sensors: dandelion-inspired microflyers, which we call dandidrones. Released in swarms, these insect-scale drones will monitor air quality, pollutant dispersion and the health of remote ecosystems with an endurance, resolution and cost that no existing drone can achieve. The challenge. Microflyers fly at Reynolds numbers of ca. 100 to 1000, where viscous and inertial forces are comparable, wakes are unsteady, and the classical tools of aircraft aerodynamics fail. Furthermore, a microflyer is pushed around by every gust. Therefore, to design a dandidrone, one needs to predict how a porous, filamentous, freely falling body responds to an unsteady wind, and how this response can be harnessed to stay airborne and to steer. The project. You will develop high-fidelity Computational Fluid Dynamics (CFD) simulations of freely flying microflyers on High-Performance Computing (HPC) facilities, including ARCHER2, the UK national supercomputer hosted at the University of Edinburgh. These simulations will be accelerated by Artificial Intelligence (AI), for example through machine-learnt surrogate models and neural operators trained on the CFD data. Successively, you will distil the physics into reduced-order models grounded in theoretical fluid mechanics, such as vortex-ring and unsteady aerodynamic theories, that predict the flight of a microflyer in a gust in seconds rather than days. Finally, these models will be used to design dandidrones with longer flight times, passive stability and controllable trajectories. The balance between numerical and theoretical work will be tailored to your interests. A unique wind tunnel. Your predictions will be tested in our accelerating wind tunnel, the only wind tunnel in the world where one can measure the response of untethered bodies, such as microflyers, to transverse gusts. In fact, you will see your models fly. A world-leading team. You will join a team of about a dozen postdoctoral researchers and PhD students working on experimental, numerical and theoretical fluid dynamics. The team is funded by the European Research Council (ERC) Consolidator Grant DANDIDRONE: A Dandelion-Inspired Drone for Swarm Sensing, and by the Advanced Research and Invention Agency (ARIA) grant SEED: Self-degrading Environmental Exploration Drones. You will work alongside colleagues who build and fly prototypes, measure flows and develop theory, and you will co-author papers across these activities. The team's recent work has been published in Nature, Nature Communications, Science Advances, Journal of Fluid Mechanics and Experiments in Fluids. What you will gain. You will be trained at the interface of fluid mechanics, AI and HPC, three areas in high demand in academia and industry. You will present your work at leading international conferences, such as the APS Division of Fluid Dynamics meeting and the European Fluid Mechanics Conference, and you will collaborate with our partners in biology, materials science and robotics in the UK, Europe and Asia. Finally, you will do this in Edinburgh, a UNESCO World Heritage city and one of the most vibrant in Europe. Who we are looking for. We welcome applicants with a first-class or upper second-class degree, or equivalent, in mechanical or aerospace engineering, physics, applied mathematics or a related discipline. Experience in fluid mechanics, programming, CFD or machine learning is an advantage but not a requirement. Instead, we value curiosity, mathematical aptitude and the ambition to open a new field of research. Applications from candidates of all backgrounds are encouraged.

 

1. Cummins, C., Seale, M., Macente, A., Certini, D., Mastropaolo, E., Viola, I.M. & Nakayama, N. 2018. A separated vortex ring underlies the flight of the dandelion. Nature 562, 414–418. https://doi.org/10.1038/s41586-018-0604-2 This paper reveals how the dandelion flies. The pappus, a disc of about 100 filaments that is more than 90% empty space, generates a separated vortex ring: a vortex that is detached from the body yet remains stable and steady above it. This is a new class of fluid flow. The porosity of the pappus is finely tuned to stabilise the vortex, and the resulting drag per unit area is about four times that of a solid disc. The discovery is the foundation of the dandidrone concept and was highlighted in an accompanying Nature commentary (https://doi.org/10.1038/d41586-018-07032-6). 2. Seale, M., Kiss, A., Bovio, S., Viola, I.M., Mastropaolo, E., Boudaoud, A. & Nakayama, N. 2022. Dandelion pappus morphing is actuated by radially patterned material swelling. Nature Communications 13, 2498. https://doi.org/10.1038/s41467-022-30245-3 This paper shows that the dandelion changes its shape in response to humidity. When wet, the pappus closes, which modifies the separated vortex ring and reduces dispersal. The closure is actuated by the radially patterned swelling of tissue at the base of the filaments. Therefore, the dandelion is a passive, morphing flyer that adapts its aerodynamics to the environment, a principle that can be transferred to dandidrones. 3. Yang, J., Bhattacharyya, S., Potnis, A., Zeng, H. & Viola, I.M. 2026. Optical manoeuvring of dandelion-inspired fliers with vortex-enabled stability. Science Advances 12, eaee8014. https://doi.org/10.1126/sciadv.aee8014 This paper presents a 1.2-mg dandelion-inspired flier whose six filaments are bent by independent light-driven actuators made of liquid crystalline elastomer. By changing the shape of individual filaments with light, the flier can hold altitude, flip, form patterns, swarm and follow controlled three-dimensional trajectories. Particle image velocimetry reveals a stable asymmetric separated vortex ring, which provides passive stability and enables steering. The flier descends at ca. 0.5 m/s, as the natural dandelion, with about 45% better positional stability. Hence, the vortex that keeps the dandelion airborne can also be used to control artificial microflyers. 4. Viola, I.M., Potnis, A., Bhattacharyya, S., Williams, E.J., Halley, D. & Murphy, D. 2025. An accelerating wind tunnel for testing untethered bodies in transverse gusts. Experiments in Fluids 66, 65. https://doi.org/10.1007/s00348-025-04135-5 This paper presents our accelerating wind tunnel, the only wind tunnel in the world where the response of untethered bodies to transverse gusts can be tested. Conventional wind tunnels test bodies that are held by a support, which constrains their motion. Instead, in this facility the body flies freely while it encounters a controlled gust, and its trajectory and the surrounding flow are measured. The PhD student will use these measurements to validate the simulations and the reduced-order models. 5. Pandi, J.S.S., Gungor, A., Bose, C., Attili, A. & Viola, I.M. 2026. Effect of transverse gust on free-falling plates. Journal of Fluid Mechanics 1040, A3. https://doi.org/10.1017/jfm.2026.11850 This paper investigates, through two-way coupled fluid-structure interaction simulations, how two-dimensional plates falling freely in a fluid respond to a transverse gust. The gust acts as a transient horizontal force that displaces the plate relative to the air, generating circulation and a lift whose upward component slows the descent. In all the cases tested, the gust temporarily reduced the descent velocity and thus increased the time aloft, with the altitude gain peaking at a density ratio of ca. 15. Therefore, free-falling bodies can scavenge energy from wind fluctuations. These insights underpin the reduced-order models of the present PhD project.

Further information

Background. Battery-powered drones remain airborne for only a few tens of minutes. Hence, they cannot sample the atmosphere continuously over large areas. Nature offers a different solution. In fact, plant seeds are dispersed by the wind over tens to hundreds of kilometres without consuming energy. The dandelion is a champion of this strategy: its pappus, a disc of about 100 filaments, is more than 90% empty space, yet it generates more drag per unit area than a solid disc. This is due to a separated vortex ring, a stable vortex that hovers above the pappus and that we discovered in 2018. The dandelion therefore shows that a microflyer can be passively stable, light, cheap and long-lived. Dandidrones, the microflyers designed by our team, aim to replicate and extend these capabilities: each dandidrone will carry a sensor of a few milligrams, and swarms of hundreds will map the atmosphere at a fraction of the cost of current technologies. Scientific challenges. The aerodynamics of microflyers is in its infancy. Three questions remain open. First, how do porous and filamentous bodies generate lift and drag in unsteady winds, when the flow both passes through and around the body? Second, how does a freely falling body respond to a transverse gust, and can this response be designed to gain altitude rather than lose it? Third, how can a microflyer be steered with the minimum actuation, exploiting the stability provided by its vortex wake? Answering them requires simulations that resolve the flow around hundreds of filaments while computing the free motion of the body, and theories that capture the physics in a few equations. Methodology. The project comprises three tasks. (1) AI-accelerated CFD. You will perform high-fidelity CFD simulations of freely flying microflyers in steady and gusty winds, using immersed-boundary and fluid-structure interaction solvers running on HPC facilities, including ARCHER2. You will then train AI models, such as surrogate models and neural operators, on these data to explore thousands of designs and wind conditions at a fraction of the computational cost. (2) Reduced-order models. You will develop low-order models of the flight dynamics based on theoretical fluid mechanics, including vortex-ring dynamics, unsteady aerodynamic theory and the equations of motion of rigid bodies in a fluid. These models will reveal the physical mechanisms behind gust response and stability, and they will be fast enough to simulate swarms and to control dandidrones in flight. (3) Validation and design. Your simulations and models will be compared with measurements in our accelerating wind tunnel, which generates transverse gusts on untethered bodies, and they will guide the design of the dandidrones manufactured by the team. Research environment. You will be based in the School of Engineering at the University of Edinburgh, in the group of Prof. Ignazio Maria Viola, Professor of Fluid Mechanics and Bioinspired Engineering. The group includes about a dozen postdoctoral researchers and PhD students funded by the ERC Consolidator Grant DANDIDRONE and by the ARIA grant SEED. Within the team, experimentalists measure the flow and the flight of models in our wind and water tunnels, numericists develop CFD and AI tools, and theoreticians develop low-order models. You will interact daily with all of them. The team collaborates with leading groups in plant biology, smart materials and robotics, and with the Edinburgh Parallel Computing Centre (EPCC). Training and career. You will receive training in CFD, HPC, machine learning and theoretical fluid mechanics, and in transferable skills through the University of Edinburgh's doctoral training programmes. Our alumni hold academic positions and roles in aerospace, renewable energy and technology companies worldwide.

Closing date: 
The separated vortex ring of the dandelion (EXP) and of a permeable disk (CFD) Apply now

Principal Supervisor

Assistant Supervisor

Funding

Funding is available for this project for Home and International students, which covers tuition fees and a stipend.

Further information and other funding options.

Informal Enquiries