BOSTON — Lockheed Martin and Boeing Co. are squaring off in a competition to showcase an experimental antenna that could expand the coverage of a planned constellation of radar reconnaissance satellites near the end of the next decade.

The Pentagon awarded $19.5 million contracts to each of the companies May 18 to continue design work on the Innovative Space Based Radar Antenna Technology (ISAT) program.

Northrop Grumman Corp. of Los Angeles, whic h is competing against Lockheed Martin to build the Space Radar satellites, was knocked out of the ISAT competition. Bob Bishop, a Northrop Grumman spokesman, declined to comment.

The Defense Advanced Research Projects Agency (DARPA), which is funding the ISAT work, plans to choose a single contractor in 2006 to conduct a space demonstration in 2009 or 2010. DARPA is seeking $45 million for the ISAT program in its 2006 budget request, and plans to ask Congress for $43 million in 2007.

The agency’s 2005 budget for the work is $46 million.

The ISAT program envisions the development of a satellite antenna that could be placed in a launch vehicle’s payload shroud in a package about the size of a sport utility vehicle. After reaching space, the antenna would unfold to its full length of about 300 meters. The space demonstration would use a reduced-scale version of this antenna, which would be about 100 meters long.

The Air Force currently plans to begin launching Space Radar satellites into low Earth orbit in 2015. The satellites are planned to provide high-resolution imagery to intelligence users and detect moving targets on the ground for the military, regardless of weather conditions or time of day.

The service had planned in 2002 to begin launching those satellites in 2010, but that date has been pushed back several times due to congressional reductions to the Air Force’s budget request for the program.

Congress has been reluctant to fund the Space Radar program, having canceled a two-satellite demonstration in this area called Discoverer 2 in 2000, and directed the Air Force to take the Space Radar program off the acquisition track last year and focus on research and development work. Members have repeatedly expressed concern on the cost of a Space Radar constellation, which the Air Force estimated in 2004 as costing at least $34 billion.

The Air Force came back to Capitol Hill with an aggressive plan for the program in February, requesting $226 million for Space Radar in 2005, adding a small-scale space demonstration in 2008, and replacing the colonel who led the program with a brigadier general who is slated to receive his second star.

The House of Representatives cut $100 million from the Air Force request for the program in its version of the 2006 Defense Authorization Act, with staffers calling the service’s plan for the program unacceptable. The Senate Armed Services Committee trimmed $75 million in its version, which is awaiting a vote on the Senate floor.

Defense Department officials have coveted the idea of putting reconnaissance satellites in medium Earth orbit because it would enable coverage of a larger area with fewer satellites. However, defense officials also have been concerned that the technology needed to provide surveillance from a higher orbit would not be ready in time for the initial Space Radar constellation.

DARPA’s budget justification materials given to Congress in February said the use of an antenna based on the ISAT vision could enable continuous coverage with 10 to 12 Space Radar satellites in a medium orbit, while about 96 would be needed for continuous coverage from low Earth orbit.

The Pentagon has recognized that 96 satellites would be unaffordable, and has indicated that a Space Radar constellation in low Earth orbit likely would involve fewer than 10 satellites that could be connected to aerial reconnaissance assets to expand the coverage area.

The higher orbit also reduces the relative motion of the spacecraft to that of the ground, easing the task of detecting moving targets, according to Michael Gross, director of radio frequency space programs at Waltham, Mass.-based Raytheon Co., which is providing the payload for Boeing.

The ISAT antenna also could be used in a higher low Earth orbit than that originally planned to give it the ability to detect and track airborne targets like aircraft or cruise missiles, according to Jerry Sobetski, ISAT program manager at Lockheed Martin Space Systems in Denver, Colo., which is leading a team that includes Harris Corp. of Melbourne, Fla., as the payload provider in the competition.

Tracking airborne targets would require about 12 to 14 satellites, according to DARPA’s budget justification materials.

Air Force Brig. Gen. John “Tom” Sheridan, director of the Space Radar program office in Chantilly, Va., said that his office is watching the ISAT work closely and may incorporate the ISAT technology on future blocks of Space Radar satellites if the ISAT demonstration is successful.

The companies competing for ISAT have been working on a series of ground demonstrations of the deployable antenna over the past year, Gross said. These tests have examined how to deploy the structures and how to calibrate the sensor to compensate for possible antenna movement in space, he said.