Bjorn’s Corner: Synthetic Vision

By Bjorn Fehrm

By Bjorn Fehrm

29July 2016, ©. Leeham Co: In last week’s Corner we discussed the power of the eye versus other senses of the human body. If one provides the eye with a convincing visual scenario, it can override many other senses that tell the brain another story.

When flying in bad weather, there is nothing in the human body that helps us to say which is up or down. Gravity should do it, you say, with the inner ears balance organ and the bums pressure situation telling us if we are up or down. Not so sure! It is very easy to slowly enter a yawing downward spiral which produces a perfect 1G force straight downwards in the aircraft, telling our brain we are doing just fine.

Birds can’t fly in clouds for the same reason we can’t, at least not without aids. The standard aid is the artificial horizon, Figure 1.

Horizon free

Figure 1. An aircraft’s artificial horizon. Source: Sensorworks Android phone app

I choose the picture because it shows the problem with an artificial horizon well. Tell me what is happening in the picture? The orange bits in the middle is the aircraft. Is the horizon leaning to the right or the aircraft to the left?

One can gradually intellectually understand that a horizon does not lean so the aircraft is rolling left. But it is not a very intuitive tool to understand what is happening. Let’s now see what happens when we make the horizon more real. 

In Figure 2 we have the same situation as in Figure 1.

Garmin synthetic vision

Figure 2. A Synthetic Vision artificial horizon. Source: Garmin

Now you tell me if the horizon is leaning right or the aircraft is on a (slightly downward) left turn. For anyone who has trained in an aircraft, there is no doubt. The aircraft is in a descending left turn. It is not intuitive that the pictured horizon should be leaning; it is against all we know since we could see.

Figure 1’s horizon is less clear. It does not reference anything we have experienced that we could see. It requires intellectual thinking and lots of training to always be interpreted correctly. This training shall also work when we are in panic mode because things have gone pear shaped. We know from the accident reports it doesn’t work out in too many cases.

Synthetic Vision

The example shows why the addition of a picture of the surrounding world is so important when flying in bad weather. It doesn’t have to be the real world seen through the windows/canopy. It can be a synthetic copy which is good enough and presented on the Primary Flight Display (PFD) or the Head Up Display (HUD).

The avionics companies have been working on this for the last 20 years and the first Synthetic Vision system was offered on the Gulfstream Business Jet in 2009. Today, virtually all avionics manufacturers offer it as part of their Primary Flight Displays.

It is used for Business jets and General Aviation aircraft. Figure 2 is actually Garmin’s iPad flight planning app (Electronic Flight-bag app) where one can put the iPad in the aircraft and couple it to a cigarette pack sized ADS-B receiver which has an artificial horizon reference built in. The app plus receiver including functioning Synthetic Vision based PFD presentation/database cost around $1,000. Add an iPad mini and you have spent $1,500 to get complete pre-flight planning support and Synthetic Vision during your flight with all the maps, charts and weather you need. There are several suppliers of EFB that offer this combination, ForeFlight EFB with Stratus receiver is another example.

So not only is Synthetic Vision available, it’s available at bargain cost levels. The example is only usable for light aircraft but for transport class aircraft there are certified avionics solutions for less than $15,000.

Not in airliners

The functionality I described is available and used in General Aviation aircraft and Business jets. It is used there to improve the pilot’s situational awareness. He has a much better grip on where he is and how the terrain looks around him with Synthetic Vision instead of the classical artificial horizon display.

Yet I have not seen it in a commercial airliner. Why? The presented Synthetic Vision is a system which is non redundant and has no self monitoring to tell the pilot when he can no longer trust what he sees. This is required for using Synthetic Vision PFDs as a certified landing aid.

With this included, it leads to improved minima for ILS and LPV approaches. Both Honeywell and Rockwell Collins have such systems and they have been working with the authorities for years to certify them. But no commercial airliner aircraft OEM has introduced Synthetic Vision into their cockpits?

Why? It’s available and used elsewhere in aviation. We will dig deeper into why no airliner is equipped with Synthetic Vision in next week’s Corner.

20 Comments on “Bjorn’s Corner: Synthetic Vision

  1. Hi Bjorn!

    While the first example is more abstract and presents less information it is also quite clean and clear.

    The second example presents a marvelous amount of information with graphics but do you find that this can be counter productive when it gets to a certain point of info overload?

    For instance the information strips for speed and altitude, might it be safer and more efficient to just indicate the speed and perhaps an arrow pointing out the direction of change. It would free up a lot of screen space and maybe make things a little easier to pickup in a glance.

    Do these synthetic displays have options for the arrangement of information and what information is displayed?

    I know you are a fighter pilot and are used to ingesting and processing large amounts of information quickly but might this possibly become information overload for a low hour general aviation pilot?

    Just curious for your thoughts…

    • The PFD picture you see is pretty much standardized re what information to show and how and where to show it. One needs it all and get used to read it as shown. But it’s all software controlled and could be de-cluttered according to the situation at hand.

  2. Personally, I’m more concerned with tunnel vision than synthetic vision!

    If, for example, in AF447, the pilots had lost all instrumentation but the artificial horizon and altimeter, would they have lost the aircraft?

    Is there a case for, in the event of mis-matched signals or reversion to direct mode – the avionics systems briefly reverting to show only the core readings required to fly an aircraft.

    Horizon, compass, altimeter, airspeed and throttle.

    In the case of AF447, airspeed was mismatched, hence would be blanked.

    • It was all there, including airspeed. it was only gone for a short while. The problem was that the co-pilot did not fly on the artificial horizon, for whatever reason.

      • Sorry, I don’t think I was clear in what I said.

        If the pilots had only the artificial horizon and the altimeter, would they have been able to avoid getting confused and thus not lost the aircraft?

        The horizon and altimeter rely on a bubble and static pressure – both of which can be made very tolerant of conditions – unlike the dynamic part of the pitot tube which iced up on AF447.

        Is there a case for “in emergency, pare instrumentation back to the bare minimum until the pilots hit an override/a time period elapses/the FCS registers the aircraft in a safe flight condition”…?

        • “If the pilots had only the artificial horizon and the altimeter, would they have been able to avoid getting confused and thus not lost the aircraft?”

          Yes. If the pilot had just kept the aircraft level and the altitude right, he could have estimated adequate airspeed by the amount of engine power he had applied – and everything would have been OK. But…he didn’t do any of that.

          What happened to Air France Flight 447 bothers me…Really Bothers me. Not because it’s unique, but because it’s more common than we’d like to admit. Despite all the preaching on the importance of instrument flying, a lot of pilots either don’t really “believe it”, or just don’t “get it”, and revert back to “Seat of the Pants” flying at the wrong times.

          I could be wrong.

  3. One of the interesting questions about synthetic vision is how to avoid it becoming ‘too compelling’. If the scene presented on the PFD or whatever display is very realistic, there is a danger that the pilot becomes fixated on the synthetic vision and stops going ‘head up’.

    There’s also a question about whether you present TRK or HDG.

    • You need the Synthetic picture both head down and up i.e. on the HUD.

  4. Being a non pilot they both look wrong to me-there’s a challenge for your abilities of explanation
    Another thing that I wonder about from looking at videos is the way the artificial horizon seems to bounce around in the same way as if it was a traditional compass surrounded by alcohol. This seems to open the possibility of confusion in a stressed pilot.

    • Accept that it’s not easy to grasp after my explanation. In real life the artificial horizon does not bounce, must be a camera effect.

  5. Well, from the Commercial OEM standpoint, Flight Deck Strategy is all about what the commercial operators are willing to pay for in terms of training. Bizjets and GA are generally much more willing to pay for new training. Commercial operators, much less so.

    In today’s nickel and dime environment, every new product has to have a certain level of commonality with the previous iteration (some so significant that the new product only brings a new packaging, with little to no new functionality). We dream of a life that the Bizjet OEMs have where they are able to bring in all kinds of current technology into the flight deck, either through their own design or partnered with their prime supplier for the instrument panel package.

    It is interesting that you pick on the artificial horizon, and only the Western style of Artificial Horizon. What about the Russian style where you have a fixed horizon in all but the pitch display function, but the airplane bug rolls left and right. Would that be more intuitive or less for an untrained eye?

    • I have never flown a Russian style horizon but I don’t like that the horizon would not be aligned with the real horizon.

  6. maybe I’m talking stupid, but I’ll say it anyways. I used to help design and integrate displays (Multi Function/Multi Purpose) on military aircraft and it became apparent to me real quick that while displays were fine, I could, as a pilot, probably live just fine without an elaborate all-glass cockpit. What I figured out was that the HUD was “God: for about 90% of what needed to be done and it didn’t require the pilot to take his eyes off the mission. Everything from stores to waypoints to artificial horizons could be displayed on the HUD – and let the pilot be a lot more efficient. Furthermore (and it’s been a few years since I’ve seen them) I don’t why you couldn’t also integrate the Radar Display into the HUD and toggle between that and your primary Flight Screen as needed.

    Those HUDs are amazing.

    I could be wrong.

  7. Figure two looks pretty good, I remember the old days training under the hood, I always felt like I was turning, during straight and level flight. Very easy for the senses to get confused.

    It occurs to me that this could be a part answer to unexpected upset accidents, helps the pilot orient himself quickly esp when time is of the essence, not applicable with AF447 but plenty of other cases.

    • MatinA: It sounds like you were more susceptible to vertigo.

      I never had it occur under the hood but did have it in real IFR.

      It went away, concentrated on the instrument and believed them and while milady annoying, it was not an issue.

      Its a matter of training and experience. Nothing is going to make it go away, its an ear balance thing, but you can discipline it into a non problem.

  8. Why is it the horizon line that rotates Bjorn?

    If I was designing a UI for stressful situations I’d always go for something that would work with instinct. So I’d have the key reference plane (ie horizon line) as the non-rotating element, with the ‘wings’ overlay then giving an instinctive corrolary to old style yokes, car steering wheels etc.. Also, the blue/green shades choice in Fig 2 is good and design not too cluttered, whereas the shades choice, cotrast and layout clarity all seem poor in Fig 1.

    • Actually you have to think yourselves in the aircraft. The horizon line in Fig 1 and 2 stay level to the outside real horizon, the aircraft is banking. If you let the aircraft move relative to the rest of the aircraft you will have problems when the pilot can see both the outside world and the artificial horizon at the same time.

      • Basically that was my question as well. I think l understand the answer although maybe I should try and blag a go on a simulator to fully get my head around it!

  9. Bjorn: While I have no issue with synthetic vision, or with more accuracy I think its a fantastic idea and innovation to supplement instruments (not replace) .

    Some of this is a bit of a rant.

    That said, I had no problem with the AI either, as a matter of fact I loved it.

    I can immediately tell you just looking at the picture, the aircraft is turning left and has a nose down attitude and is likely loosing altitude (instrument scan and or cross reference the VSI and altimeter to confirm)

    I never had any interpretation of it other than the orange wings and dot represents the aircraft and its relationship to the horizon. If you don’t get that you should not be piloting.

    And I have been in more than one situation where what appears to be straight and level (or any variation) of terrain is level is indeed not, its an optical illusion.

    If AF447 pilots had looked and knew how to interpret theirs along with the VSI and altimeter that accident would never have happened.

    My only issue with it was that it would tumble (gen av aircraft) if you really went wild, not sure anyone could recover from that on the AI but while I don’t think I would have gotten there, it would be a chance.

    Having seen some of the new displays that have an overwhelming amount of information, I have to wonder if I could have learne3d to fly those vs the steam gauges that worked so well if you knew your craft and trade.

    I think slow and careful is a very good idea, unfortunately all of this has its flaws as well and there is no replacing good pilotage. It is possible to make a bad pilot look good until push comes to shove (as see in too many cased)

    I believe that all equipment added to the flight deck should have a valid confirmation that it works in the real world with average pilots.

    Per Boeing and their auto throttle, get it confused and results are very bad.

    Bells and whistles and complaining betty get ignored, so why have them when all it does is add to chaos.

Leave a Reply

Your email address will not be published.