Avoiding Collisions

Flying high over Baltimore last week, two Martin 404 airliners hurtled toward each other on what seemed like a sure collision course. Seconds before disaster, the planes suddenly veered apart. One swept upward; the other dipped slightly. So coolly had the maneuver been executed that it seemed as if an invisible electronic hand had guided the controls.

Both pilots, in fact, did have a helping hand. The "near miss" was the first public demonstration of a promising new collision-avoidance system (CAS) that may reduce some of the risks of flying in the nation's increasingly crowded skies. Last year the U.S. had 38 aerial collisions, a 46% rise over 1967. In the years ahead, the risks will increase, as more planes--including jumbo jets and SSTs (see BUSINESS)--join the rapidly growing U.S. air fleet.

Successive Alarms. The new system was sponsored by the Air Transport Association as insurance against airborne chaos. Like the planes used in Baltimore last week, each aircraft equipped with CAS is, in effect, shielded by a huge protective electronic bubble. When one plane's bubble brushes another's, it triggers successive alarms in both cockpits. The first comes 42 seconds before the moment of collision as calculated by the CAS's onboard computers. If the planes are still headed toward each other at 30 seconds, a flashing red light warns the pilots to prepare for evasive action. Five seconds later, the computers issue their final command. Depending on the relative positions of their planes, one pilot may be ordered to climb, the other to dive or stay level.

The secret of the system is timing. To form the electronic bubbles, each participating plane must send brief radio pulses--none longer than a tiny fraction of a second--in an assigned order of rotation at exact three-second intervals. The system demands such accuracy that all the planes must carry atomic clocks, which are precisely synchronized to a master timepiece on the ground or aboard one of the planes. Theoretically, CAS is so fast and efficient that it can safely handle as many as 2,000 planes over an area of more than 61,000sq. mi.

The ATA, as representative of the nation's airlines, would like to install the system (estimated cost: $24,000 to $50,000 per unit) aboard all commercial aircraft by 1974. But there is one serious drawback. Unless CAS is also carried by private planes, it will not prevent such collisions as the one between a big passenger jet and a small private plane near Indianapolis last month that killed 83 people. Many aviation men feel that the only long-range protection against more aerial tragedies lies in an all-encompassing, new air-traffic control system that would keep tabs on every plane in U.S. skies.

Pulsing Satellites. A preliminary blueprint for just such a scheme was presented last week by California's TRW Inc., a major producer of electronic equipment. Dubbed "Navstar," it would rely on at least four orbiting satellites to serve as aerial traffic cops for the nation's air lanes. Unlike ATA's device, which depends on radio pulses from the aircraft, the satellites themselves would beep at a precisely controlled rate. Tuning in on these signals, an airborne computer could get an exact electronic "fix" on the plane's position and altitude. By measuring shifts in the frequency or "pitch" of the radio waves as the plane moved toward or away from a satellite, the computer could also calculate the aircraft's speed.

Most important, the process could be reversed. If the plane itself transmitted a readily identifiable signal, it could be picked up by the satellites, relayed to a tracking station on earth, and fed into air-traffic-control computers. By comparing the relayed signals, the ground-based equipment could tell traffic controllers the exact location, altitude and speed of every plane aloft. The necessary onboard transmitter, as TRW points out, would cost no more than a few hundred dollars--easily within the means of almost every private plane owner in the U.S.

Though aviation officials think that air-traffic control satellites are at least a decade off, Navstar's designer, Electrical Engineer David D. Otten, is more optimistic. If work started immediately, says Otten, a Navstar-type system could be put into service in only three or four years.

This file is automatically generated by a robot program, so reader's discretion is required.