Almost all of us have wished at one time or another to be a fly on the wall-but nobody wants to get swatted. A safer option might be to command a fleet of intelligent flies that could seek out and report critical information while we stay safely removed from danger. With that idea in mind, engineers at MIT’s Lincoln Laboratory are now developing a micro air vehicle-a semi-autonomous spy plane small enough to hold in the palm of your hand.

The idea of producing the world’s first miniature intelligence-gathering plane was proposed three years ago at Lincoln Lab, where researchers sought a way to provide direct access to reconnaissance data for soldiers serving in small military units, such as those deployed in urban settings. They envisioned a portable surveillance system that could quickly inform soldiers of imminent, unseen dangers. In urban areas, for instance, such a system could enable soldiers to “see over the hill and around the corner,” says William R. Davis, who manages the Lincoln Lab micro-air-vehicle program. Advanced versions might sniff out nuclear, biological, and chemical weapons in hostile terrain, assess battle damage, or monitor hostage crises or Waco-style standoffs.

Based largely on concepts outlined at Lincoln Lab, the Defense Advanced Research Projects Agency (DARPA) launched a $35 million program this year to develop prototype micro air vehicles, soliciting preliminary proposals from industry and academia. Organizations competing for funding include university laboratories such as Georgia Tech Research Institute (GTRI), aerospace companies, and small businesses.

At the same time, DARPA has provided initial funding to Lincoln Lab to develop a fully functioning prototype, which researchers expect to complete within three years. Weighing two ounces and measuring less than six inches in length and width, the prototype vehicle will fly at 20 to 30 miles per hour, operate within a radius of up to 3 miles (the limit relates to the expected range of the vehicle’s communication system), and remain airborne for up to an hour. It must also have reconnaissance and navigational capabilities.

“By and large,” says Milan Vilajenik, who heads Lincoln Lab’s Engineering Division, “this is going to be a flying chip.”

Given its tight size and weight constraints, getting this flying chip off the ground and keeping it there will be no mean feat. “As you go down in size,” Vilajenik explains, “existing technology is too big. Most of the subsystems will have to be developed.”

The first challenge is to create an efficient wing design that can provide enough lift and low enough drag for a vehicle in this size range, in which aerodynamic behavior differs from that of larger, faster aircraft. The Lincoln Lab team calculated that a hovering craft with rotor blades would require about twice the power of a fixed-wing vehicle. To minimize power requirements, therefore, the team is evaluating several fixed-wing configurations using propellers for propulsion, says program manager Davis.

But Robert J. Englar, a principal research engineer at Georgia Tech, argues that even with a propeller, a conventional wing will not generate enough lift to keep a very small, slow vehicle moving through the disruptive airflow that it will likely encounter. Georgia Tech has submitted a proposal that Englar declines to discuss, but Sam Blankenship, coordinator of GTRI’s Microflyer Program, says that engineers may ultimately need to look to the flapping wings of small birds and insects as models for energy-efficient flight.

A strong propulsion system can compensate for shortcomings in aerodynamic performance, notes Davis, but more powerful propulsion units are likely to weigh more-and designers want to reserve weight for data-collecting sensors and communication systems. The Lincoln Lab team determined that at the small scale required, jet engines would consume too much fuel, while power sources such as batteries would weigh too much and provide too little power. Lincoln Lab engineers identified internal-combustion engines and fuel cells as the most promising near-term alternatives and hope to create miniaturized versions of internal-combustion engines within one to two years.

Along with a robust propulsion system, the diminutive vehicle needs a flight-control system so that it can maintain its course in the face of air turbulence or sudden gusts of wind. Because the plane will travel out of the troops’ sight and may encounter rapidly changing flight conditions, “a soldier can’t fly the vehicle like a model airplane,” says Davis. His team determined that to execute maneuvers, the prototype could rely on small-scale devices-sensors that measure aircraft speed, acceleration, and atmospheric pressure, and electrical actuators that move the plane’s aerodynamic surfaces.

Davis points out that sophisticated microfabrication techniques make it possible to manufacture sensors and actuators with low power requirements on a very small scale. As microplane development matures, however, designers expect to replace these tiny devices, which use moving parts and must be mounted separately, with micro-electromechanical systems-high-precision systems that resemble computer chips and are produced using methods similar to microcircuit fabrication. These could be embedded in a micro-plane’s wing, saving precious weight and providing more efficient control.

Finally, Lincoln researchers plan to develop a very small imaging system for the vehicle and a portable ground station-a laptop computer and a small parabolic communications dish-to transmit photos. They envision a two-gram, one-cubic-centimeter visible-light camera positioned beneath the plane and obtaining million-pixel images-pictures sharp enough to identify military vehicles and personnel from a 100-meter altitude. The challenge in developing the imaging system is not to design or manufacture individual components but to integrate them without exceeding tight mass and power constraints.

Indeed, integrating this tightly packed assortment of reconnaissance, propulsion, flight control, and other subsystems will be the final hurdle, Vilajenik notes. For example, designers may have to isolate a vibrating, heat-producing internal combustion engine from an imaging system sensitive to those disturbances, and prevent electromagnetic interference between electric motors and the communications antenna.

As engineers develop ways of squeezing more functions into smaller packages over the next decade or two, they expect that payload capacity, endurance, and range will rise. The Lincoln Lab team envisions advanced micro air vehicles that perform tasks as diverse as detecting chemical signatures of unconventional weapons, deploying stationary sensors to monitor unpatrolled areas, and imaging and recording the sounds of scenes in and around buildings.