The Future IS The Sky of Unmanned Planes

The vast majority of air crashes are caused by human pilots’ operational errors. The increasing level of aircraft automation can better solve this problem.

The Future IS The Sky of Unmanned Planes.
The Future IS The Sky of Unmanned Planes.

Air crash caused by human error

On June 1, 2009, Air France Flight 447 encountered a tropical storm in the middle of the Atlantic Ocean on its way from Rio de Janeiro, Brazil, to Paris, France. A few minutes later, the Airbus A330 crashed into the ocean, killing 228 people on board. Four years later in July, on a sunny morning, a Asiana flight was flying to San Francisco airport when one head hit the breakwater at the end of the runway. the tail of the flight broke and the broken fuselage rolled over the runway. The accident resulted in 3 deaths and dozens of injuries.

On the surface, there is no correlation between the two accidents, but on closer inspection, they have another sad commonality: the pilots of both planes believe that the flight control system can automatically prevent the plane from stalling in mid-air, or prevent the plane from crashing due to its too slow speed. They are all wrong. On February 4, 2015, Taiwan’s Fuxing Air Crash killed more than 40 people. The black box is interpreted as: it does not rule out the pilot turning off the wrong engine to cause the crash. The research found that this kind of man-machine mixed operating system is an important reason for many air disasters, and this kind of confusion is becoming increasingly serious. With more and more automatic components, pilots may feel confused when the plane has a serious failure, unable to figure out when automatic driving will end and when their duties will begin. The result was a plane crash (Figure 1).

So, is it time to completely eliminate pilots? This is obviously what many airlines think. They argue that rather than letting pilots stay in the cabin to drive, it is far better to transfer them to offices thousands of meters away for remote supervision and let the plane fly automatically. In terms of safety, this is indeed a reasonable plan, because about half of all fatal flight accidents are caused by the mistakes of crew members.

If safety is strengthened, unmanned passenger planes can greatly reduce the expenses of airlines and passengers. Professor Mary Cummins, who studies automatic flight at Duke University in the United States, pointed out that without pilots, airlines can save a lot on salaries, simulators, training, medical insurance, hotels and pensions, thus reducing ticket prices. Moreover, automatic flight should save more fuel, which will further reduce costs and reduce greenhouse gas emissions. In addition, without the crew members in the nose, the passengers in the front row can enjoy the magnificent panorama-of course, there will be an extra charge. There is only one key question left: how would you feel about boarding an unmanned plane?

The Problem of Unmanned Aircraft

At the heart of this revolution is a simple fact: a great deal of work on modern planes is done by computers, and airliner pilots have little chance to drive. The automatic flight computer will take control of the plane just 30 meters from the ground and continue to drive according to the speed, heading and altitude entered by the pilot into the flight management system. And long ago, computers were able to track the radio beacons on the runway and pilot the plane to land automatically.

However, automation also brings new difficulties to pilots: when automatic flight computers encounter conditions outside the program, such as structural damage to the aircraft or extreme weather like Air France Flight 447, they will suddenly throw the driving right back into the pilot’s hands. In the event of such a serious accident, the pilot may be at a loss as to the sudden arrival of the right to drive. The same danger is that the on-board computer will also pour a large number of warning signals, checklists and audible alarms to the pilots at critical moments, keeping them overwhelmed. For example, in November 2010, the engine of Australian Airlines Flight 32 carrying 469 people exploded, cutting off 650 control circuits. However, when Captain Richard de Crespi tried to land in Singapore, he and his co-captain were dazzled by the 120 command menus flashed on the screen.

The US Federal Aviation Administration has also realized that the combination of human and computer is not necessarily the safest. In 2013, the U.S. Air Traffic Control Administration released a seven-year research report that listed the dangers posed by the combination of human and airborne computers. Casey Abbott, chief investigator of the U.S. Air Traffic Control Administration, found that pilots often rely too much on automatic flight systems and “are reluctant to intervene even if they suspect that the system is not functioning properly.” The report also warns that pilots are still confused about the status of the automatic flight system.

In the short term, airlines are going to change their computer systems and strengthen training for pilots to reduce these risks. In the long run, lowering the level of automation in the cockpit seems to be a good way. However, many aviation practitioners are pursuing another path, which is to further reduce the role of error-prone humans in the cockpit.

Boeing, Airbus, BAE Systems and Dassault have jointly invested 30 million euros to develop a computer-aided system to help pilots cope with emergencies in four years. This project is called “Advanced Cockpit to Reduce Pressure and Workload” (ACROSS). It aims to reduce the dangerous peak value in the pilot’s workload by using computers to avoid chaos in an emergency. The purpose of this system is to enable the captain or the deputy captain to deal with the crisis alone when the other party is unable to move. However, its function does not stop there. Its designers hope to establish an automatic flight system that can completely replace the co-pilot, so that only one pilot is needed for the passenger plane from takeoff to landing. One of the functions of ACROSS is to recognize the difficulties that need to be overcome to achieve this goal.

However, even if highly intelligent machine intelligence can really serve as a qualified co-pilot, what will happen if the only pilot loses his mobility? In Cummins’s view, in order for the computer assistant captain to drive with human beings, it must be able to control a plane alone when the captain is unwell and complete the take-off, cruise and landing tasks on its own. If that is the case, what else do we need human beings to do? The elimination of human beings will definitely reduce the chaos in man-machine cooperation.

Is human being redundant?

In 2013, NASA’s Ames Research Center supported this reasoning in a study. The study pointed out that the automatic flight system installed in the new generation of single-person light jets will also increase the burden on pilots and make mistakes in navigation and flight control. Cummins said: “All flight researchers admit that the reduction of pilots from two to one is only an intermediate step and the ultimate goal is to have no pilots.” Not only that, Cummins also believes that the era of fully automatic flight will come sooner or later. When she was a U.S. navy pilot, she understood one thing: the fighter’s automatic flight system didn’t need her participation at all, and its technology of landing aircraft on aircraft carriers was far superior to that of human beings. She said: “It is much faster than me in adjusting direction, airspeed and altitude.” The operation of computers in a very short period of time is indeed better than that of human beings, because the communication speed between human eyes and brains is not as fast as that between sensors and processors. “From when you see a stimulus that requires action to when you actually take action, it takes at least half a second. In contrast, it takes only a few milliseconds for a computer to detect a signal and take action, which is absolutely beyond the reach of human beings. ”

The experiments of U.S. military unmanned aerial vehicles have also proved this point-with the improvement of automation, their crash rates have also decreased greatly. In the past, accidents often occurred during take-off and landing, but the situation improved obviously after the two operations were handed over to the computer. Cummins pointed out: “Landing and taking off are almost the heaviest tasks for pilots.” Almost half of the fatal air crashes occurred during these two flight phases.

Unmanned passenger planes have appeared

So, how far is the drone from us? In fact, it has already appeared: a BAE Systems unmanned jet has already flown 800 kilometers to test its interaction with other planes and air traffic controllers-but the plane is still equipped with a pilot as a precaution.

This jet is equipped with the preliminary machine intelligence that future UAVs will need-an intelligent flight management system called “airspace integration processor” (AIP). AIP uses radio, radar and camera sensors to prevent collisions and communicates with a ground pilot via satellite. Rod Buchanan, BAE’s flight engineer, said AIP would draw up a route after receiving the destination coordinates, taking into account dangerous weather conditions and air traffic control.

Despite all the progress, it is unlikely that airlines will start pilotless aircraft first. Freight companies such as FedEx and UPS will take the lead in reducing the number of pilots from two to one. This is equivalent to testing the technology on an aircraft of the same model as the airline, except that there are no passengers on board. If the test is successful, these cargo planes will completely abandon the pilots around 2035. The company only needs to keep a pilot and let him monitor different cargo planes on the ground by satellite and control them remotely in case of failure.

Now it seems that the success of automatic flight technology may be greatly advanced. Thanks to the rapid development of electric propulsion and intelligent flight control systems, researchers believe that the era of flying cars or “private air vehicles” (PAV) may be coming soon. Their success in business depends on whether they can be fully automated. After all, learning to fly costs thousands of pounds. If the public wants to fly safely into the crowded air in PAV, we need a computer to drive. Engineer Mike Janp pointed out: “If we move the scale of ground traffic into the air, we can no longer use manual control.” He predicted that the flying car would lift off in 2025.

Automation will undoubtedly attract airline operators: besides fuel, labor is the biggest cost in this industry. Senior pilots earn more than 240,000 US dollars a year, and companies often need to train and supplement them. In addition, they have to pay for benefits such as health insurance and pension. Their working hours are also strictly limited. Donoghue Wilson, an aviation consultant at the University of Coventry in the UK, pointed out: “Crew members have physiological needs and strict work cycles, which limit flight time.”

Cummins pointed out that automation can also save more fuel-the autopilot system will not deviate from the optimal route, so it will not waste additional fuel. And technologies that are good for airlines and the environment are also good for passengers because they can reduce fares. If drones can reduce the turnaround time between flights, they may be able to reduce the flight delay slightly. Aircraft designers can even change the layout in the cabin. Wilson pointed out: “Since pilots no longer have to sit in front, designers no longer have to follow traditional designs.” When the cockpit is cancelled, passengers can enjoy the scenery that only pilots could see before.

Public Psychology of Fully Automatic Driving

Of course, fully automatic driving also has a big hidden danger: if the system fails and there are no crew members on the plane, what should we do? In January 2009, chesley Sullenberg’s plane had an engine failure over new york. His plane made a forced landing on the Hudson River, with no casualties on board. An artificial intelligence “captain” could not have made such strange and clever decisions.

Brent Walker Smith, who studies risk and technology at the University of Southern California, Columbia, said computers may not be able to do this, but airlines may already be prepared for such special events. He said: “If automation can reduce the number and extent of crashes and casualties, then even if the planes produced are more likely to crash, their actual risks may not be higher.” However, pilots certainly cannot willingly be eliminated by new technologies. De Crespi, for example, pointed out: “Only by copying human consciousness, perception and prediction into one machine can we produce reliable unmanned aircraft. Until then, the hope of saving passengers will only fall on the pilots. ”

Richard Toomer, spokesman for the British Aviation Pilots Association, also agreed: “Passengers want to put their lives in the hands of two fully trained and rested pilots. This will not change immediately. ” However, as driverless cars are about to hit the road, the public may be more receptive to driverless vehicles-Cummins said she has seen some evidence in a recent study. However, she also admitted that drones would still be rejected by many people. “It is easy to reduce the number of pilots from two to one, and difficult to reduce from one to none.” She said, “This technical problem is subject to regulatory and cultural resistance.”

Robert Bohr, a clinical psychologist at the Royal Free Hospital, pointed out that this cultural resistance has deep roots. Bohr’s research area is aviation psychology. He pointed out: “This resistance stems from the primitive fear of losing control of things.” Passengers like to see a competent person controlling the plane, a person who is in danger like them when there is an accident. Although the pilots admitted that they were only monitoring the automatic system, Boer believed that passengers still regarded them as “professionals dressed in uniform and calm”, and they “operated the plane on our behalf and fought against danger”. Bohr said, “The aviation industry needs to prove that the safety of unmanned aerial vehicles is the same as or even better than manned aircraft. I need extensive experiments to clearly prove that it is indeed safe. If someone told me tomorrow that the pilots had all gone to work on the ground, I would also feel worried. I think it is not easy for people to accept this. ”


Original Essay,Writer:AC SMART,wanna repost ? Please sign the references:

Leave a Reply

Your email address will not be published. Required fields are marked *