Ads-b and sailplanes

On Monday, April 6, 2015 at 11:36:33 AM UTC-7, wrote:

Sorry, FLARM is not an "anti-collision" system like TCAS, it is a "traffic advisory" system like ADS-B. The only thing that FLARM-specific algorithms do is reduce the number of warnings provided of nearby gliders that are determined not to be on conflicting paths. Having identical firmware in all FLARM units simplifies the programming issues and allows for use of lower-powered processors. Implementing similar algorithms on top of the more diverse ADS-B environment will not have any innate tendency towards making matters worse.


I'll explain.

After talking to the Flarm engineers, I concluded that you need to have a more nuanced view than "anti-collision" and "traffic advisory". The middle ground is what Flarm does and ADS-B does not. Call it "collision advisory" then. It tells the pilot not only is there an aircraft within some airspace volume around him, but does it have a trajectory (forecasted out some tens of seconds and based on knowledge of what type of aircraft it is) that has a reasonable probability of intersecting his aircraft's trajectory. It then displays quadrant and altitude differential for the pilot to act on (TCAS additionally tells the pilot what to do, that is true, but it doesn't fly the airplane, so the pilot is still in the loop for collision avoidance). Collision advisory like Flarm must account not only for filtering threats from non-threats, but also how pilots are likely to respond to how the threat is presented. Flarm uses a particular convention, but others are possible - all the way up to TCAS-type RAs.

The filtering of collision threat versus non-threats in a glider scenario is the entire difference in suitability between Flarm and ADS-B. Trying to do this kind of filtering without a position forecasting algorithm is challenging because you can't effectively filter threats from non-threats. Without a common position forecasting algorithm across all systems you might have one system give collision advisories based one one set of assumptions and another system issuing advisories on another set of assumptions and could lead to "you zig, I zag" kinds of asymmetric warnings and pilot reactions.

Consistent warning algorithm and advisory display results in better (but not perfect) pilot response to collision advisories. Think about a head-to-head approach. It's hard enough for pilots to deal with Flarm today (turn left? turn right? climb? dive?) but it can get worse if the warnings come at significantly different times (or not at all) and with different philosophies of what to display. Consistent position forecasting and consistent display of the threat are critical to pilots taking successful and non-conflicting action to avoid a collision.

To illustrate - Imagine three different (ADS-B-based) systems on three different gliders in a thermal - one forecasts turning and uses total energy based on received GPS position and trends (this is not as effective or reliable as doing what Flarm does by forecasting future position prior to transmission, but leave that aside for now), another algorithm does a straight extrapolation of the current instantaneous velocity vector and a third is position only. Imagine the first system also gives you an indication of where the threat will be relative to you when it is expected to be at closest approach so you can steer away from that point and the second just tells you when either gilder is pointed at the other and indicates where to look to pick it up. The third one will give you advisories continuously as long as any other glider is within a thousand feet of you or so (my LX 9000 does this, but only once when it picks up a new target. Even so, I want to turn it off most of the time). In a crowded thermal these three different approaches to collision threat filtering and display could - with pilots in the loop and reacting - create all kinds of chaos that I would think of as making matters worse.

That's why I conclude not having a single collision advisory algorithm and display philosophy could make matters worse. I wonder if this is why ADS-B systems are pretty conservative about what they tell the pilots about traffic vs collision threats.

Andy