When ADS-B traffic uplink was announced, there were great expectations for what it could do to improve safety, specifically, to reduce mid-air and near mid-air collisions. After some years of flying with ADS-B traffic, my expectations have been, shall we say, down-sized. It’s nice to think that improvements are easy, but there are real world constraints.
Please note that while some features of some products are discussed below, this article is about ADS-B traffic displays in general and is not a product comparison. And if I got some product details wrong…
Expectation 1: Reducing Midair Collisions
There’s a lesson from thirty years ago that ADS-B traffic display designers seem to have missed. When the airlines got TCAS (Traffic Collision Avoidance System) in the late 80s, the hope was that it would solve the problem of midair collisions. While TCAS has indeed saved lives, one of the lessons learned was it wasn’t enough to just toss hardware into the cockpit and expect the pilots to figure out what to do with it. TCAS was not a stand-alone system, rather, it was part of the larger system of airplanes, airspace and ATC. In particular, when TCAS gave resolution advisories (commands), both flight crews needed to follow TCAS and ignore any ATC commands. Eventually that became standardized, but there was at least one mid-air collision in which one airplane followed TCAS and the other followed ATC.
ADS-B now gives traffic data but the question is, what are pilots supposed to do with that data? ADS-B does not give resolution commands, nor do ADS-B cockpit displays interact with ATC. When ATC asks the pilot to report traffic in sight, that means visual, not ADS-B. There is no pilot/controller terminology for ADS-B. Is ADS-B just another system tossed into the cockpit that the pilots are supposed to figure out on their own what to do with?
There are new regulations requiring drones to have position reporting equipment on board. It remains to be seen if the TCAS mistake will be repeated yet again, with drones showing up on cockpit displays as raw data with the expectation that the pilot will have time to see it and figure out what to do with that data.
Expectation 2: Aid to Spotting Traffic
In my airplane, I can display ADS-B traffic on two different kinds of screens. One is analogous to a radar scope. Fully zoomed in, the screen will show runways, but otherwise, no background at all. This means that the display doesn’t help the pilot look for a target over a specific area, like on a final approach path.
Worse, that radarscope-style display shows all traffic around the airplane. Traffic behind the airplane is often irrelevant, and dedicating screen space to unimportant traffic means less screen space for important traffic. Yes, there is the case of traffic overtaking from behind that needs to be considered, but the key measure of threat traffic is not how far away it is in distance, but how far away it is in time. ForeFlight, for example, displays traffic based on distance and on time. I’ve read that in Australia, arriving traffic calls state minutes to the airport, not distance.
Other kinds of cockpit screens that can show traffic have different kinds of map backgrounds, but those map backgrounds can have enough background clutter that traffic symbology can be hard to spot. Zooming in on the display seldom increases the size of the traffic icon, so the targets are still hard to spot.
Especially in the vicinity of an airport—where collisions are more likely—what would be best would be a background designed specifically to help in spotting traffic. That background should show airports and runways, extended runway centerlines, and maybe a very few prominent ground features like rivers, maybe not. This is a different concept from tossing ADS-B traffic icons onto an existing map not designed for spotting traffic.
Yes, when a map is shown and is primarily being used as a map, showing traffic on that map is useful. However, that’s a different case from a screen designed primarily to show traffic.
It’s not that easy, though. At the high end, there may be standards that displays must conform to, and some vendors may have compatibility to maintain across all product lines. I think this is a solvable problem, but then, I’m not in charge of implementation or convincing the regulators.
Expectation 3: Aural Alerts
A human factors challenge is what and how much information should be conveyed by voice and how much by a warning sound. That topic is far beyond the scope of this article, but the bottom line is that voice alerts are too compelling to use casually in the cockpit as they can distract the pilot from other tasks. Also, voice alerts can and have blocked radio communications. Recall that TCAS (airline level Traffic Collision Avoidance Systems) give voice notifications but on the first Traffic Advisory say only, “Traffic, traffic,” not all the details. And TCAS Resolution Advisories are similarly terse. Sometimes less is more, especially when it comes to voice alerts.
Expectation 4: All Traffic of Interest is Displayed, Part 1
Many ADS-B traffic displays display traffic only within a selectable altitude band. For example, jets thousands of feet up in the flight levels are appropriately not displayed. This altitude band concept works satisfactorily in cruise flight for us low performance GA types, but it doesn’t work well around busy airports. Here’s why.
Around an airport, an ADS-B display should show all traffic of interest, i.e., most all traffic.
When I’m in the pattern at my local airport with Class C airspace, ADS-B won’t show incoming fast jet traffic until it’s within the altitude band selected on my traffic display, even though that fast jet traffic is already talking to the tower. Sometimes ADS-B shows traffic talking to the tower, sometimes it doesn’t. Sometimes I can count on ADS-B to show me all the traffic of interest, but sometimes it won’t.
The “solution” is for the airborne ADS-B system to look for fast moving traffic descending within, say, 20 miles of the airport. If such traffic appears, the ADS-B system could display that traffic with the reasonable expectation that the traffic is inbound to the airport and will soon be within the normal altitude band.
Here’s the rub: not all the airliners have ADS-B Out. If they don’t have it, the ADS-B uplink system will only send traffic within 3500 feet of my altitude, only slightly larger than the 2700 foot plus/minus altitude “Normal Band.” It would still be worth it, says I, for ADS-B equipped jet traffic, like bizjets. That data could come via ADS-B air to air.
Yes, it is possible to reset the altitude band of the ADS-B near the airport, but that’s extra work at the wrong time and would show much more traffic than is of interest.
Expectation 5: All Traffic of Interest is Displayed, Part 2
A few months back, I was in cruise flight, fat, dumb and happy with no traffic on the ADS-B screen. ATC called and reported traffic at three o’clock, altitude unknown. Sure enough, over at three o’clock, at my altitude, was an older plane, possibly not even transponder equipped.
I had started to get into the habit of expecting ADS-B to show me all traffic, but… If there is threat traffic around, I want ADS-B to tell me about it, and not to disregard it if that plane is not ADS-B equipped.
Expectation 6: The Screen Will Always Be Readable and Meaningful
Here are some screen shots taken over the past few years:
Here’s why predictor arrows necessary! Pilots don’t have enough time to watch the screen to determine relative velocities and collision threats.
Better, but… some of these are irrelevant if the pilot continues straight ahead, especially if groundspeed shows that my plane can’t catch them or that they can’t catch me. Those irrelevant targets increase workload and potential for error. Note how much of the screen is reserved for traffic behind, which is not a factor. Some targets are filled in, some are outlines. That too-subtle distinction does not reduce screen clutter. In the traffic pattern, however, it can be useful to see planes behind you.
Can you find the traffic? There are two planes shown…
This map display is useful as a map but not so useful as a dedicated traffic display, particularly when zoomed out this far. Some targets are not threats and are distracting clutter, meaning more workload to interpret the display. Note that airport and VOR labels are, for purposes of tracking traffic, meaningless clutter. These targets are displayed with motion relative to own ship, a Garmin patent. (The symbology on this screen shot is not consistent with the most recent Pilot’s Guide.)
Here’s what a good traffic display background might look like, but this is not a default screen, this is a manually decluttered ForeFlight screen on a tablet.
To revert to a “regular” map background, all the map symbology has to be turned back on manually, one feature at a time. ForeFlight has no concept of stored screens, unlike many panel-mounted avionics. Note the use of color to show threat levels. But at Boeing, a rule was to never use color alone to convey information, and this display violates that rule. Also note that target aircraft motion is displayed relative to the ground, not relative to own ship.
A full traffic pattern and then some at Prescott, AZ. No wonder I’m hightailing it out of there…
Expectation 6: ADS-B Traffic Improves Traffic Awareness
Sometimes no, sometimes yes.
Sometimes ATC will call out traffic by aircraft type, distance and relative bearing. ADS-B will display an N-number, so I’ll ask ATC if that traffic matches the N-number on the ADS-B display. That won’t meet ATC’s requirement for visual identification, but it will let me get benefit from the ADS-B display.
And again, there is no terminology with FAA-approved significance to communicate to ATC that I am tracking traffic on ADS-B.
Here’s an idea: suppose active N-numbers were stored in the airborne ADS-B units, paired with aircraft type? That would reduce some of this N-number / make / model Q&A. This wouldn’t necessarily be perfect, of course, with an obvious example being Skyhawk versus Cessna 172. Space limitations on the display might require terse identification, like C172. Another possible error source is changed N-numbers. If research shows that problem to be significant, maybe N-numbers could change on the same 28 day cycle as chart data. But in any case, the key to success will be to make the information pilot-centric and not to just throw more information over the fence for the pilots to figure out what to do with it.
There’s hope, however. The current Rev B ADS-B standard is being replaced with Rev C, which will allow four times as much data to be sent from the airplane. No surprise, there are hundreds of proposals on what to do with that extra payload, as it’s called, including weather downloads from the airliners, or maybe even from little planes like mine. And the new payload can carry multiple kinds of messages. How? Avionics engineers often package data as a label that says what kind of data is being sent plus the data per se.
It is a reasonable expectation that the new ADS-B systems will be backwards compatible with existing systems, but older systems might not be able to send data in the new format—like aircraft type.
Expectation 7: ADS-B Lets Me Fit In Better With Traffic Flow
There are really three classes of traffic that ADS-B can help the pilot with: immediate threats, potential threats, and non-threat aircraft that are traffic to fit in with.
Sometimes ADS-B helps me fit in with other traffic, sometimes not. When there is jet traffic landing at the same airport I am, the controller has a plan that I gladly fit in with, most of the time. Sometimes I get excessively large vectors that seem to be more for controller comfort than for any reason obvious to me. And there have been times when I’ve been asked for maximum speed, followed shortly thereafter by minimum speed. Hey, no problem, I’m a big boy.
What is useful, though, is when I’m on final behind a local flight school Cessna 172. When I see their groundspeed, I can make sure that I’m not closing in on them. They might be slow clearing the runway because a student pilot is flying. Again, no problem. Or if they’re flying downwind on the other side of the runway, I can adjust speed to fit in behind or ahead of them.
Similarly, when I’m waiting to take off, occasionally I can tell the tower I can accept an immediate takeoff if the ADS-B shows me the traffic on final. How? From the groundspeed of the traffic on final, I can easily calculate their approximate rate of descent (rule of five) and use that with their altitude to calculate how long till touchdown. In my plane, from clearance to take off to crossing the other runway is 60 seconds. Sometimes, though, it’s an exercise in frustration when an excessively conservative controller slows things down.
Meeting More Expectations for Usability
- There are significant increases in ADS-B traffic display usability to be easily achieved with cockpit screen displays designed to aid in visually spotting traffic.
- Aircraft type could be added to ADS-B displays. Screen real estate could impose limitations on how many characters are used for the aircraft type, but this idea seems at first blush to be valuable for matching ATC callouts with ADS-B traffic. This idea is one of many candidates in the upcoming new ADS-B data protocol.
- Arriving high speed traffic could be displayed around airports, even outside conventional altitude bands. This would work for jets with ADS-B Out, but not all airliners have ADS-B Out.
- There may be good reasons why ATC does not care whether a pilot sees traffic on ADS-B or not. If those reasons have been publicized, that publicizing has been inadequate. If there are no reasons, ADS-B traffic could be utilized to improve things for both ATC and pilots.
- All traffic should be uplinked to ADS-B equipped traffic, even non-transponder traffic. Not sending traffic information to pilots “just because” wastes ADS-B as a safety resource.
- If it hasn’t been done already, the FAA should figure out what pilots are supposed to do with drone traffic on an ADS-B display. Inundating the pilot with raw data can degrade safety, not enhance it.
Acknowledgement: Thanks to Garmin media relations for access to their ADS-B guru for technical information.