Autopilot on Boeing and Airbus?

[V8LM]

Olivero said:

First, thank you to everyone who replied. It is very much appreciated.

We are a small team based in New York and France. At the moment we are working on the next generation of super computers who can quickly crunch huge amounts of data. Think of IBMs Jeopardy, but with raw data rather than quiz questions.

Our idea is to improve on the (very) short term decisions auto pilot uses. Weather effects, eg wind could be compensated for in real time. Imagine that the flight path was updated many times every second. Our aim is not to replace any of the existing systems, but rather to build an add-on that would create an optimized flight path. The biggest two benefits we imagine would be fuel savings and a smother ride.

It is too early for us to show a prototype, but not to have general discussions.

Any significant update to a flight path "many times per second" would be a nasty ride.

[trashbat]

djc206 said:

Conflict detection software is now very good, I don't think that's really the biggest obstacle.

The collision avoidance & separation minima bit of conflict resolution is only part of it. It's also about load on sectors (and the human controllers), avoiding SUAs like military exclusion areas etc, resourcing at airports etc.

It's an interesting domain with a load of scope to make changes, but probably quite rightly it's also very conservative and it'll take a very long time for anything to be implemented. And you don't play around with the real world, hence big money in modelling & simulation.

Incidentally, OP, the company I worked for was a small business based in Paris and Maryland, working with FAA & Eurocontrol. I maintain occasional contact so drop me a PM for their contact details if you think there might be some mutual opportunity.

Edited by trashbat on Thursday 27th November 11:11

[djc206]

ATFM is a dark art!

[rhinochopig]

Olivero said:

rhinochopig said:

Oli, are you familiar with IEC61508? Your team will probably need to be.

Yes, very familiar...
Our system will not be mission critical. It would operate in addition to existing systems rather than any kind of replacement. A simple flip of a switch could turn it off without effecting autopilot.

Your claim of not safety critical depends. IF the system is providing information to the operator on which they may base safety critical decisions, then it can be change its class. And SIL2 and above is VERY expesnive to justify for software based systems.