A version of this article appears in the August 25 issue of Aviation Week & Space Technology.

NASA’s long-running push to find a commercial route for crews to the International Space Station takes a big step this week, with the selection of a company or companies to take the design work they have done with government seed money and move on into space. Call it a wild guess, but I anticipate we will all be watching crew capsule flight tests over the next few years. Two of the three contenders—the Boeing CST-100 and the SpaceX Dragon—are capsules, and previous funding decisions have favored those designs over Sierra Nevada’s Dream Chaser lifting body.

Even if NASA selects Dream Chaser for the Commercial Crew Transportation Capability phase of its public-private crew vehicle development project, the agency will need to flight-test its Orion crew capsule. Dream Chaser already has flown a free-flying glide test marred only by a landing gear failure after a successful approach and touchdown, and the company may tow it behind a C-17 for future tests in the atmosphere (AW&ST June 12, p. 31). In that sense, its flight-test regime up until launch to orbit will be similar to that of the space shuttle orbiter, which deployed the Enterprise atmospheric test article from a modified Boeing 747.

But testing requirements for capsules differ from those of the shuttle, which added ejection seats to the orbiter Columbia for John Young and Robert Crippen on its first spaceflight, and carried pilots on the Enterprise flights. The Dragon is already flying unmanned cargo missions to the ISS, and Boeing plans an unmanned Atlas V flight of its CST-100 capsule before sending a two-person crew to orbit.

All three commercial crew contenders have hired retired shuttle astronauts to oversee planning for flight-crew operations. At Boeing, Christopher J. Ferguson is director of crew and mission operations.

Ferguson, who commanded the final space shuttle mission—STS-135 on Atlantis in 2011—was a Navy test pilot before joining NASA, and flew the shuttle to space three times. He has been deeply involved in preparing his company’s commercial crew vehicle for flight-test. But unlike shuttle flight, CST‑100 pilotage will be largely autonomous.

The idea is to make it repeatable and safe, and you do that by employing a lot of autonomy, Ferguson says.

But capsule pilots won’t be sitting idly by while machines do all the work.

“We’ve been levied the requirement to ensure the pilot can take over at any time,” says Ferguson. “We will plan this mission for months in advance, right down to every translational maneuver, every rotational maneuver, when power-up events occur, when power-down events occur. We anticipate [unexpected occurrences] that may require us to sort of interrupt the autonomy, correct it, and then reengage the autonomy.”

Boeing maintains a cockpit simulator (see photo) at its commercial-crew facility in Houston, which has been upgraded using rapid-prototyping as the design is advanced. One area getting a lot of attention lately is how the vehicle-monitoring system in the Atlas V will trigger an abort to get the CST-100 and its crew safely away from a failing vehicle. Along with construction of a launchpad crew access tower that will include a slide-wire emergency escape system and an enclosed “white room” for ingress; the emergency detection system is one of the next steps in CST‑100’s development.

“For those time-to-criticality failures that are very short—rapidly degrading situation, loss of control—there will be an automatic signal,” Ferguson says. “You’ll exceed a redline inside the emergency detection box; it will send a signal to the CST-100 to initiate an abort.”

The capsule carries a pusher-type hypergolic-fuel abort system that can empty its tanks in about 5 sec., subjecting the crew to about 6g as it boosts the capsule away from a failing launch vehicle. If it isn’t used, the propellant remains available for maneuvering once the capsule reaches orbit. The mission control center also will be able to send a command-abort signal to the vehicle, and the crew will have its own abort control.

“We can enable all three, or perhaps go down to one,” Ferguson says. “But regardless, the range safety system always trumps the emergency detection system. We can never completely turn it off for obvious reasons.”

Following a pad-abort test at White Sands, New Mexico, by the end of 2016, the unmanned Operational Flight Test on an Atlas V is planned to dock autonomously with the ISS. The vehicle would remain at the station for two weeks under current plans, and then return to a dry-land touchdown on parachutes and airbags in a western desert landing zone. Results of that test would be incorporated into the first flight-test with a two-person crew—from Boeing and NASA—officially slated before the end of 2017.

However, while Boeing has not made a formal decision, John P. Mulholland, vice president and program manager of commercial programs for its space exploration unit, says it is unlikely the company will continue if it does not receive at least some funding in the upcoming contract.