Robot vehicles modeled after insects. By Joe Conroy and Andrew Hyslop.
The notion that small, even very small, UAVs might have practical uses arose in the early 1990s. In 1992, DARPA conducted a workshop titled "Future Technology-Driven Revolutions In Military Operations". One of the topics in the workshop was "mobile microrobots". The idea of using very small "microdrones" was discussed, and after initial skepticism the idea started to gain momentum.
The RAND Corporation released a paper on the microdrone concept in 1994 that was widely circulated. DARPA conducted a series of "paper studies" and workshops on the concept in 1995 and 1996, leading to early engineering studies by the Lincoln Laboratories at the Massachusetts Institute of Technology (MIT), and the U.S. Naval Research Laboratory (NRL) in Washington DC.
The studies demonstrated that the concept was feasible. In 1997, DARPA then began a multi-year, $35 million USD development program to develop "micro air vehicles (MAVs)". The MAV project's goals was to develop a microdrone whose largest dimension was no more than 15 centimeters (6 inches); would carry a day-night imager; have an endurance of about two hours; and be very low cost. It would operate with a high degree of autonomy to be used in the squad-level combat environment. MAVs capable of hovering and vertical flight would be used to scout out buildings for urban combat and counter terrorist operations. A MAV could be included in a pilot's survival kit. A downed pilot could use it to keep track of enemy search parties, or relay communications to search and rescue units.
More : http://en.wikipedia.org/wiki/Miniature_UAVs