The ignominy of spending years wringing out a vehicle designed to eliminate human input still raises Volk’s ire. He flew the full-size atmospheric prototype, which was outfitted with four 18,000-pound-thrust Lyulka jet engines, on more than half of its 24 flights. The first landings were manual. Then, over successive flights, Volk gradually relinquished control to the auto-land software, until the last few landings were accomplished by computer alone. At that point, Volk says, his interest in the Buran project waned.
For the people who worked on it, this was another of the program’s bitter ironies. The first contingent of nine cosmonauts assigned to Buran in 1977, a year before NASA hired its first shuttle astronauts, included some of the most accomplished test pilots in the Soviet Union—and, in the case of Igor Volk, the world. But their job was to serve as understudies in a vehicle designed to require no human input.
Most of the first Buran cosmonauts were employees of the Gromov Institute, outside Moscow, Russia’s premier center for flight research, while others came from the Institute of Experimental Flight, a branch of the Ministry of Defense. Volk and Rimantas Stankiavicius, who also was picked in 1977, were the two main pilots for the series of atmospheric flights conducted between 1985 and 1988. Three teams of two pilots made 24 short-approachand-landing flights in a full-scale Buran prototype. But when the vehicle finally reached orbit strapped on the back of the giant Energia rocket, the pilots stayed home. And that may stand as Buran’s single unique accomplishment: It returned from its two orbits and landed like a conventional airplane, controlled entirely by computers. The Buran’s auto-land software, though kluged together, worked perfectly on its one and only space mission.
Sergei Krikalev became a cosmonaut in 1985 and has gone on to become perhaps the most experienced space traveler alive (see “The Captain, the Pro, and the Fighter Pilot,” Feb./Mar. 2000). He recalls watching the Buran’s flawless launch from the Baikonur cosmodrome. November 15, 1988, dawned with a storm front from the Aral Sea moving across southern Kazakhstan. Energia’s launch had been put off several times, and officials declared that this would be the day, despite the clouds, wind, and 40-degree-Fahrenheit temperature. Liftoff occurred at 6:00 a.m.
About three and a half hours later, the Buran landed on a specially built runway north of the launch pad. “As a pilot, I guess I was a little jealous when it emerged from beneath the cloud cover on its final approach,” Krikalev says, grinning. “I thought the computers would get it to, say, 4,000 meters and then it would drift away. It touched down within a meter and a half of the center stripe and stayed within a meter of the center of the runway until it stopped. It looked as though a good pilot was at the controls.” High praise from a cosmonaut who also happens to be a former world champion aerobatic glider pilot.
Krikalev had, like many of his colleagues working on Buran, worried about the accuracy of the first landing. The auto-land software had been developed at five independent organizations, each of which had written a program to command the Buran to (1) leave orbit, (2) descend to an altitude of 60 miles, (3) glide through the atmosphere at an altitude of 13 miles, (4) make its approach to a microwave- and telemetry-equipped airstrip at Baikonur, and (5) flare, get the landing gear down, and roll out. Given the potential for something to go wrong, “it was oddly satisfying to see it land as perfectly as it did,” Krikalev says.
But Victor Zabolotsky, who copiloted Buran on its last taxi test in 1989 and is now president of the Russian Federation of Amateur Aviators, disputes the common claim that the landing was successful, saying that the landing approach the computer chose at Baikonur was “stupid, with a high percentage of risk. Most amateur aviators would not have considered that approach.” An animated speaker with deep creases above his right eye and a scar over his left eye, Zabolotsky says that on a scale of one to 10, he would give the Buran landing a four. “There was a crosswind at about 30 degrees to the airstrip that had been blowing since the launch,” he recalls. “A normal pilot would have kept the crosswind on the right wing, then simply made a left turn across the wind and an approach and landing with the crosswind on the left wing. The Buran computer chose to make a left turn over the center of the runway, then a hard descending right turn into the crosswind for final approach and landing.
“When it emerged from the clouds, it had a close encounter with the MiG-25 escort flown by [test pilot] Magomet Tolboev, who had intercepted the Buran at 9,000 meters [29,000 feet] above the cloud layer. He lost it in the clouds, then had to take an evasive maneuver under the clouds when the unorthodox pattern the computer selected flew the plane into airspace where no human pilot would have anticipated encountering it.”
Zabolotsky nonetheless admits, grudgingly, that there was a logic in using the auto-land system. No Soviet space vehicle had ever been allowed to fly with people until it had conducted two successful orbital demonstration flights. The Buran program was simply complying with a rule going back to Korolev’s days. Faced with the same dilemma, NASA had waived its own safety rule, and on the first shuttle mission, John Young and Robert Crippen rode an untested rocket. To this day, they are the only astronauts to do that.
Zabolotsky doesn’t want to concede too much, however. “There was the capability [on Buran] to override the computers in an emergency,” he says. “I believe that option would have been exercised, often. I don’t believe a pilot like Igor Volk, who could dead-stick land every plane we ever made, was going to sit idlyas a computer aims his plane at a runway.”