SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:


Print PageEmail PageView PDF

Solar & Alternative Energy

Designing solar airplanes for continuous flight

A novel conceptual design tool helped scientists build Sky-Sailor, a small-scale, autonomous solar-powered craft that flew for 27 hours.
17 June 2009, SPIE Newsroom. DOI: 10.1117/2.1200906.1649

The dream of flight powered only by the sun's energy has long motivated scientists and hobbyists. Over 30 years after the maiden flight of the first solar airplane prototype, Sunrise I,1 using an unmanned aircraft continuously day and night has now become feasible. Such airplanes store excess solar energy, typically in batteries, for night use. Zephyr, a prototype with an 18m wingspan developed by QinetiQ, recently completed a stint at high altitude that lasted more than two days.2 Our prototype, the Sky-Sailor, has a wingspan of 3.2m. It flew for 27h in the summer of 2008. We constructed the prototype with the goal of sending an aircraft to Mars (which imposes size limitations). Scientists hope such an aircraft will one day fly in the Martian atmosphere—theoretically forever—collecting data. Despite constant improvements, battery technology remains the crucial limiting factor to this endeavor, not the efficiency of the solar cells.

In order to assess the feasibility of continuous flight, we developed a generic conceptual tool at the Autonomous Systems Lab (ASL) to optimize the main design parameters such as wingspan, aspect ratio, and battery mass with different technological and mission states.3 We aimed to reduce iterative, expensive, time-consuming design loops by directly providing a design parameter set.

Figure 1. Airplane weight as a function of wing area, S, and aspect ratio (AR) for both existing planes and fitted model.

First, we studied power considerations. The efforts began with maintaining equilibrium: the energy obtained from the sun must be at least equal to the energy spent for level flight. Despite lower gravity on Mars, the power consumption for continuous flight in its lower atmosphere would be generally higher than on Earth, due to its very low air density. Therefore, improvements on the battery energy density are required. (Current lithium-ion cells achieve approximately 240Wh/kg.)

Next, we analyzed the airframe. Finding an appropriate scaling law for the airframe is very challenging and sensitive compared to all of the other components. In addition, in our model the airframe weight is currently expressed solely as a function of wing area and aspect ratio (AR) derived from data of existing gliders and model aircraft (see Figure 1).

We assembled the models, which allowed us to analyze the design space for continuous flight. Figure 2 shows the time the plane could fly in excess after the night (safety margin). It is shown as a function of wingspan and total mass varying with the choice of battery size. Notice that the plot is only valid for a distinct parameter setting. Clearly, a wingspan just over 3m seems most favorable.

Figure 2. Excess endurance of a solar airplane designed for continuous flight. Main settings: AR 12.6, minimal payload, 70% of the maximum solar irradiation on the longest day at 47° latitude.

When designing a solar airplane for daytime flight only, however, (such as for use in disaster scenarios) the energy balance constraint drops, and either the range or endurance can be maximized, as shown in Figure 3. Note that the amount of sunlight available dramatically changes the optimal configuration.

Figure 3. Solar airplane for daytime flight: endurance dependency on AR and battery weight (m_bat). Main settings: 1.8m wingspan, considerable payload, full irradiation on a summer day at 47°latitude (left) and in the darkness (right).

Using our design constraints, we built a prototype (see Figure 4).4 It corresponds well with the prediction from the conceptual design: at a 3.2m wingspan and an AR of 12.6, its total mass is 2.5kg, yielding a nominal speed of approximately 30km/h.

Figure 4. Sky-Sailor prototype.

The Sky-Sailor incorporates navigation and control electronics, allowing fully autonomous flight. The solar module is made of thin, slightly flexible silicon cells, with an efficiency of 17%, covering 75% of the main wing. In June 2008, Sky-Sailor flew for 27hr (constantly at low altitude), proving its capability of continuous flight.

We presented the main concepts for designing solar airplanes for various scales and mission scenarios. These considerations led to the design of Sky-Sailor, a small prototype unmanned aircraft that demonstrated continuous flight at low altitude. In the future, we will refine the conceptual design tool, in particular the airframe weight model, which will allow us to optimize and downscale solar airplanes towards a 1m wingspan. These planes could be used as completely autonomous remote sensors, for example, spending many hours constantly airborne in disasters.

André Noth, Stefan Leutenegger, Roland Siegwart
Swiss Federal Institute of Technology (ETH)
Autonomous Systems Lab (ASL)
Zurich, Switzerland

André Noth completed his masters in micro-engineering at the Ecole Polytechnique Fédérale de Lausanne (EPFL) in 2004 and recently finished his PhD at ASL. His work covers the design of solar-powered airplanes for continuous flight. His major interests lie in the field of multi-disciplinary design, solar-powered robots and mechatronics systems design.

Stefan Leutenegger received his MS in mechanical engineering in 2008 and is currently pursuing a PhD in robotics and autonomous systems with a focus on designing an autonomous solar-electric aircraft.

Roland Siegwart has been ASL director since 2006 and is a renowned specialist in the design and intelligent control of autonomous systems. He has a diploma in mechanical engineering (1983) and a PhD in mechatronics (1989) from ETH Zurich. He spent a year as postdoctoral fellow at Stanford University and later worked as research and development director at MECOS Traxler AG. In 1996 he was appointed professor for autonomous microsystems and robots.

Walter Engel
Engelberg, Switzerland

Walter Engel received the certification as a mechanical engineer/designer in 1966 at the Institute for Professional Development, Machine and Metal Industry Zurich. His career from 1963 to 1998 at Oerlikon Contraves Pyrotec AG ranged from engineer to manager of ammunition projects in the development department. In 1982, he also started developing electric RC-model airplanes. He achieved several first ranks in the development of extremely light RC-models for long duration flights.