Mr Atkinson (Flight International, 18-24 February) appears to have misunderstood our research, reported by Michael Phelan (Flight International, 4-10 February).
We are redesigning the flight-control laws on the fly, rather than redistributing the effort from a fixed set of control laws to the surviving control surfaces. For example, in the 1992 loss of an El Al Boeing 747-200 Freighter, the problem was not just malfunctioning surfaces, but also changed lift characteristics of the starboard wing, and we are compensating for that.
I agree that if the status of each surface is immediately available, then the kind of reconfiguration described by Mr Atkinson is appropriate and can be done quickly. But if it is not, or if other changes in aircraft behaviour need to be inferred from flight data, then some time must elapse before enough information is gathered - and this does not depend on computing power. (The computing speed is needed to make correct decisions once the new behaviour has been "learned".) The objective of our research is to cope with unanticipated failures.
J M Maciejowski Cambridge University, Cambridge, UK
Source: Flight International
