Say intentions and vatsim collaboration

Hi everyone,

I’m reaching out to discuss a potential collaboration between the VATSIM and Say Intentions developers. I’ve spoken with the Say Intentions team, and they have indicated that integrating their system to automatically detect when a pilot flying in Say Intentions enters a VATSIM-controlled airspace is an easy task. This would allow for a seamless handoff to an online VATSIM controller, letting pilots fly both online and offline simulations simultaneously. This integration would enable pilots to communicate from ground level all the way up to center and back down to ground level, even during the online portions with VATSIM.

The Say Intentions team is ready to develop this feature, but it would require approval and cooperation from the VATSIM team.

I believe this integration could significantly enhance the flight simulation experience for everyone and allow full atc, even when there’s limited online controllers.

I’d love to hear your thoughts and see if there’s interest in moving forward with this idea.

Thanks!

The VATSIM BoG is already in communications with SayIntentions.AI, see BoG MM 2025Q1 (Board of Governors Minutes).
A collaboration was also previouly discussed in 2024Q4, but dismissed at that time, you can see the relevant BoG MM for the reasoning.

Awesome

This would make best of both worlds and fill the gap of offline sectors.

Did you read the answer? :sweat_smile:

Also, determining when a pilot is in airspace that is covered by a human controller is FAR from an easy task.

1 Like

I’ve been doing that for a few years using vatspy boundary file, the aircraft lat, lon plus the VS json feed.

Ok you don’t have altitude constraints but it’s sufficient when top down is in use. It’s a feature in VSR we had to tell user if they are in controlled airspace

There are a lot of situations, even with top-down, where the VAT-Spy boundaries are not accurate. As just one example, there are many facilities that allow controllers to work only part of an FIR/ARTCC but VAT-Spy has no way of knowing that they aren’t working the whole thing, and so VAT-Spy shows the entire FIR/ARTCC lit up.

That, plus the fact that the 2D boundaries are only an approximation, due to all the shelves and special cutouts that most facilities have, makes it insufficient for determining if a flight is entering or leaving controlled airspace.

And if there is no one working the airspace top down, then the VAT-Spy data is essentially useless for this purpose. It’s fine for a tool like VAT-Spy or VSR where it’s still up to the user to interpret the data, but it’s completely insufficient for automated systems to determine airspace coverage.

It would certainly be close enough though… This is a consequenceless environment, “pretty good” is good enough.

Also though, if exact precision were desired, you’re just talking about airspace boundaries as a 3d shapefile. It’s not exactly impossible.

I find this to be a cool idea.

I have to disagree … it would not be close enough. Not if you’re trying to accomplish what the OP describes as “seamless handoff to an online VATSIM controller.” As described previously, there are way too many situations where using the VAT-Spy data would cause a handoff to the wrong controller, a handoff to the right controller at the wrong location (way too early or way too late), or no handoff when there should be one. It would “get it wrong” so often that it would regularly become more annoying than it is useful, for both the controllers and the pilots.

Don’t get me wrong, I’m not saying it couldn’t work in some places. Certainly there are some facilities where their coverage matches the VAT-Spy data quite well, and in a predictable, deterministic way, based on controller callsign and/or frequency. However, I would hazard a guess that such facilities are the exception rather than the norm.

Then you need to consider the situations where facilities do ad-hoc sectorization and splits of their airspace, like during events. There’s pretty much no way for the AI to discover these customizations.

I haven’t seen anyone claim it’s impossible, just not “an easy task” like the OP suggested.

Keep in mind it’s not just about defining 3D boundaries. It’s also about knowing how to determine which volumes of airspace “belong” to a given controller. Think about how things change when controllers sign in and sign out. Their airspace gets split off from someone else when they sign in, or combined with someone else when they sign out. Also, the airspace that a given controller owns can literally move when the runway configuration at a major airport changes.

In short, for something like this to work, we would need a way for controllers to publish the set of airspace volumes that they are currently controlling, with the ability to change that set as needed when people sign in/out or configurations change. This is something we’re slowly working towards adding in the VATUSA vNAS system, to support things like automatic handoffs (quite similar requirements to what’s being discussed here) but it’s a long ways off and it’s a ton of work to get it right.

Euroscope (and perhaps VatSys?) has support for defining airspace ownership already, so that could perhaps be used as a basis for a system that publishes airspace ownership for tools like SayIntentions.AI to use. I say this to illustrate that I’m aware that it’s possible, but far from an easy task.

I would think, until a database of shapefiles for various sectors/facilities was completed, stopgap “contact me” messages could be used as necessary, just as they are today.

Ultimately though, yes, controllers would need to define which airspace volume they’re assuming ownership of when they log on. I would think this kind of system would be desired anyway, for internal Vatsim purposes… Kind of surprised it doesn’t already exist.

Don’t get me wrong, I’m not the target audience here as I’ve never even used non- Vatsim ATC in the sim, built-in or 3rd party. I just don’t really care about simulating ATC. I like Vatsim because the human factors geek in me marvels at how all the same threats and errors we see in the real world tend to manifest in similar scenarios in the sim. Given that there are no consequences, it’s fascinating to see. But yeah, I likely wouldn’t be a SI customer, although if this kind of system ever were created I’d probably try it for the heck of it.

As I said earlier, it does already exist to some degree in some places, via Euroscope. And it’s planned for VATUSA.

Plus, for many years, various VATSIM developers have been working on coming up with a way to have an accurate database of airspace volumes and a way to allow controller clients to report which of those volumes each controller is covering. The fact that this is still a largely unsolved problem goes to show how complex it is. It’s doable, it’s just very non-trivial.

Yeah, there are plenty of ways something could be done as a stopgap. Whether or not a half-baked solution is worth the effort is another question, for which you’ll get a different answer from each person you ask. :grin:

1 Like