One contributor to the A320 crash off Brazil in 2009 (Air France 447) was that the two pilots were making opposite inputs on their control sticks. The aircraft was in a stall, and therefore it was crucial to push the nose down, to regain airspeed. The instinctive human reaction (of untrained people) is to pull the nose up, since the airplane is falling. To oversimplify a long sequence of events drastically, pilot made the correct move, but the other pilot apparently panicked, and pulled back on his control stick. He continued to do this as they fell from 40,000 feet all the way to the Atlantic Ocean.

A new accident report says that the same thing happened in the crash of an  Indonesia AirAsia Airbus A320, flight QZ8501, last year.

Critically, the captain attempted to push the aircraft’s nose down with forward pitch on his side-stick – the proper response to an aerodynamic stall.

But this nose-down input was rendered ineffective because the first officer was continuing to pull rearwards on his side-stick. These opposite actions effectively cancelled one another out, because the Airbus flight-control logic acted on the sum of the two side-stick inputs.[emphasis added]

In other words, one pilot neutralized the effects of the other. And neither pilot realized what was happening.

This situation is at least partly the result of the way Airbus designed its cockpit. Its aircraft use “side sticks”, one for each pilot. You can see the First Officer’s control stick on the right in this photo.

Sidestick controller, A320.

The two controls are not linked in any way, and unless they  look carefully the pilots can’t see the other control. So in both accidents it was possible for the head pilot (on the left) to not realize his control inputs were being neutralized. There is a digital warning of a conflict, and a procedure where the pilot on the left can lock out the other one, but apparently it takes 40 seconds of continuous pressure on a switch. 40 seconds is fine if you are at 30,000 feet, but it is an eternity at low altitude.

In comparison, Boeing aircraft have a large, ugly, and obtrusive “yoke” sitting in front of each pilot, and the two yokes are  physically connected so that they are always in the same position. If one pilot pushes and the other one pulls, they will both physically feel the conflict.

Some aircraft companies are now adopting synchronized sidestick controllers, which artificially transmit forces from one stick to the other. This will presumably help a lot.

The principal reason for using sidesticks is ergonomics. Removing the control yokes allows for larger flight displays and—in the latest cockpits—pilots can be moved closer to the instrument panels, allowing use of touchscreens. The biggest drawback of the “passive” sidesticks now used in civil aircraft is the lack of control feedback from the aircraft or the other pilot.

But the transfer of “active inceptor” technology to the commercial sector from the military is helping to overcome that objection. Active sidesticks that provide tactile and visual feedback in response to pilot and autopilot commands are used in the Lockheed Martin F-35 Joint Strike Fighter and Sikorsky CH-53K heavy-lift helicopter. ….

This ensures both sidesticks move together in response to both pilot and autopilot commands, providing crew situational awareness equivalent to conventional pilot controls, says BAE.

It’s true that a properly trained crew should not have gotten into trouble in either case, and having gotten into trouble should have been able to get out of it quickly. But modern commercial aviation did is very safe partly because it does not use the excuse “this should not be necessary” to avoid fixing known problems.  More training of crews in high altitude flying is clearly necessary, but I hope Airbus is also thinking about changing its control design.

The bigger picture: We will continue to learn new lessons about how to use sophisticated automated systems, and sometimes to “write the lessons in blood.”  Airbus’s control system design was, and to an extent remains, more radical/advanced than Boeing’s, and as such it will likely continue to have novel problems more often than Boeing.  The aviation industry as a whole is slow but very effective at learning from experiences like this, and sharing insights across vendors.