BOSTON — The U.S. Defense Advanced Research Projects Agency (DARPA) plans to begin work next year that could lead to a formation-flying experiment in which multiple small satellites perform functions normally carried out by single, larger spacecraft.

DARPA is planning to issue a request for proposals in late winter or early spring that could result in several contracts to provide concepts for at least one flight experiment under the Future Fast, Flexible, Free-Flying, Fractionated Spacecraft united by Information Exchange (F6) program, according to Jan Walker, a DARPA spokeswoman.

The F6 program has a 2007 budget of about $12 million, Walker said in a Nov. 29 written response to questions.

Several companies have done studies or other work that could be applicable , Walker said. They include Lockheed Martin’s Advanced Development Center and Northrop Grumman Space Technology of Redondo Beach, Calif., who worked under separate contracts that wrapped up over a year ago to study the concept of dividing functions traditionally handled by a single satellite among several small spacecraft, Walker said. The Massachusetts Institute of Technology in Cambridge also has participated in studies of the concept, she said.

Relevant work under way today includes studies by Payload Systems Inc. of Cambridge, Mass., and the Massachusetts Institute of Technology on a concept called electromagnetic formation flying, Walker said.

This involves keeping several satellites in a tight formation by controlling the strength and direction of magnetic fields inside each spacecraft, she said.

Space Systems/Loral of Palo Alto, Calif., meanwhile, is under contract to study a formation-flying concept in which one spacecraft equipped with solar arrays would collect and transfer power to the others, Walker said.

Electromagnetic control is among the technologies that could play a role in the F6 flight experiment, according to budget documents posted on DARPA’s Web site . Other candidate technologies include crosslinks and attitude control suited for formation flying as well as distributed systems for computing and for optical and radio-frequency data collection, the documents say.

F6 would not be the Pentagon’s first attempt to test the idea of distributing functions among several satellites in tight formations that can be reconfigured to suit the mission at hand.

The Air Force Research Laboratory’s TechSat21 experiment was designed to demonstrate the concept using three satellites flying in formation, but canceled the project in 2003 due to the technical challenges involved.

MicroSat Systems Inc. of Littleton, Colo., the TechSat21 prime contractor , leveraged some of its design work on the canceled demonstration in building the Air Force’s TacSat-2 satellite, which is slated to launch on Dec. 11.

John Roth, president of MicroSat Systems, said that in the wake of Techsat21’s cancellation, the company shifted its focus to small, flexible satellite platforms to support so-called responsive space missions like TacSat-2. But he pointed out that small, low-cost responsive space platforms work better in concert with others, and added that MicroSat likely will compete for work under the F6 program.

The relevance of formation-flying to the concept of responsive space was the subject of a paper co-written by a Owen Brown, a program manager in DARPA’s Tactical Technology Office, and Paul Eremenko, an associate in the defense business segment of Booz Allen Hamilton. The paper, presented at the Response Space 2006 conference in April in Los Angeles, argues that the most common definition of responsive space — small satellites that are launched on short notice in response to emerging tactical needs — is too restrictive.

A better definition would emphasize the ability of satellite systems to adapt in response to rapidly shifting needs among the military forces they are designed to support, the paper said.

Spreading functions across several satellites can reduce the consequences of launch and on-orbit failures , the paper said. It also facilitates replacements and component upgrades via relatively inexpensive launches, the paper said.