Situational awareness and electronics overload

It's hard not to be impressed by the latest round of navigation electronics. This is 2010, after all, an era in which the average desktop computer has the computing power to calculate the airflow around a Space Shuttle during re-entry, and we can't tell the difference between live and CGI actors on the cinema screen. I'm not convinced, though, that all this computing power is a good thing to be throwing at navigation systems- at least, not in the ways we see in some of the current crop of nav systems.

Before going any farther, I'd encourage the reader to take a look at an article by John Harries and Phyllis Nickel on knowing where it's at, and a related article by Colin Speedie on the importance of accurate information as an input to any navigation system.

I think it's fair to say that nowhere is situational awareness more critical than in the cockpit of a large airliner on final approach. So it's worth taking a look at how they do it: if you're spending a quarter of a billion dollars on a new jetliner, it's fair to expect you'll get the absolute best avionics that money can buy. Take a good look at the cockpit of an Airbus A380 (or as a pan/zoom Flash show). There are plenty of switches and buttons, of course- but look at the display screens. Notice anything unusual? This multi-million-dollar avionics package produces simple, 2D images in a small number of highly contrasting colours. The text-based displays show large, clear white or green text on black backgrounds. There is no 3D animation, no 16-million-colour shading (although the display panels are capable of this), and no fancy trim or glitz on or around the displays. The large MFDs at each side are displaying simple line drawings to convey the important nav data. And there are full QWERTY and number pad keyboards right at the fingertips of both officers.

What Airbus has figured out, and what I think the marine electronics industry ought to try, is that all that computing power is at its most useful if it is used to simplify the information presented to the captain. A good navigation computer's algorithms should take in all the available data, sort and process it, figure out what the captain is actually looking for (based on the current situation and the settings entered into the system by the operator), and display that information in a clear, concise form that can be interpreted in a quick glance. I'm not trained as a pilot (although I'm familiar with how aircraft cockpits work), but I can figure out about three-quarters of what's presented on those Airbus displays in a few seconds. Making sense of a typical NavNet 3D screenshot, though, takes me a fair bit longer- ten seconds or so just to find the "current position" marker on the 3D chart.

Another important note is that the avionics package in the Airbus A380 is designed to behave, from the pilot's perspective, in a very similar manner to its counterparts on the A320, A330 and A340. In other words, a pilot on one of those planes shouldn't have to learn a whole new computer system if he decides to re-train on another plane. I don't think marine electronics are anywhere near this level of user interface consistency yet.

I'm not against electronic navigation- frankly, I find these sorts of gadgets pretty fun to play with. I would, however, like to see all that computing power and development time put towards more important things than 3D animation and satellite imagery overlays:

  • Give us a simple display we can read at a glance. Yellow for land, blue for shallow water, white for deep water, stars for rocks, and the familiar buoy symbols. Give us a large, obvious marker representing our boat and its heading. Give us similarly obvious markers for other ships and their headings.
  • Use that computing power to distill the information into just the important bits. Emphasize things that present a collision risk, don't worry so much about ships that'll clear me by ten miles. If a buoy should be within visible range, emphasize it on the display. If I can't see or hit it in the real world, don't let it clutter up the display.
  • Give me real time, seamless pan, zoom and chart quilting. Automatically pull up the right chart at the right scale for that view. I've heard NavNet 3D can do this, but haven't had a chance to play with it yet. OpenCPN is trying, and can almost do it.
  • I want one big red button marked "MOB" or "EMRG". If I hit it, I want the system to mark that point and automatically switch to displaying a distance and bearing to that point, and to pass the relevant information to the DSC radio in case I decide to hit the Mayday button.
  • Take the money that is spent on animated menus and flashy but rarely used features, and use it to give me a solid, truly waterproof case that doesn't feel like it's going to crack open if I bump it wrong.

And, most importantly:

  • Do not try to make me or anyone else think that electronics are a substitute for real situational awareness. A computer is just a tool- in this case, a tool to bring up the right charts, AIS and radar targets at the right time so that I can concentrate on watching what's around me. If the computer regularly becomes the focus of attention for more than a few seconds, it is such a liability that I would rather do without it and pull out a paper chart (as I do now- Sunset Chaser has neither the space nor the cruising range for a fancy MFD to be worthwhile).




Situation Awareness

Great post, Matthew. I really wonder who has ever really used 3D displays. And, if they have, what fiddling around with all the options does to their situation awareness does—nothing good, I suspect.


Add new comment