All very good questions and points, which I (hopefully) address below.
BTW, please hit my website
www.potomac-aviation.com and enter your contact info.
I would greatly appreciate your feedback on its presentation of info!
---
Q1. Voice is too robotic and does not have natural breaks between words. Get a higher quality speech engine, if one is available. With Siri and Google assistant, we're spoiled with voice AI that sounds natural (mostly).
REPLY - Actually, we do modify my very real recorded voice, to be clearer and faster than natural. We found 'natural speech' was not as easy to understand across noisy cockpits.
We developed the system for super busy VHF, where limiting impact to the VHF includes keeping things as short as possible, which means speaking a bit faster.
We actually do insert spaces where a human would breathe, to make it more understandable. It's not obvious in this little clip.
The airport manager can actually adjust speech rate for local tempo. No kidding.
Funny story behind that
We also modify the voice to maximize modulation level, critical for AM VHF
The result is it does sound a wee-bit unnatural at 'full-throttle,' as used in the clip,
to make it clearer and louder than a human speaker,
but by design and for good reason.
----
Q2. Advisory is rather long. The only additional information I noticed from a typical AWOS system is the recommended runway.
REPLY - The system is adaptive and prioritizes to only give the most relevant information at the moment, adjusted for operational relevance and frequency congestion. In this example crosswind and windshear very relevant in extremis.
We call it 'SMART METAR'
The system has a number of different modes. At an airport with an approach, for example, it is usually set to emphasize visibility IF the frequency is not very busy, to answer the visibility reporting requirement.
As the frequency gets busier, even visibility disappears, depending on operational importance and real-time frequency congestion.
Q2A. If you need MicroTower to tell you the best runway, you shouldn't be PIC.
As to runway call, yes, of course we figure this out in our heads all the time. The challenge is really in more ambiguous conditions than this video, to answer the question, what are OTHER pilots already in the area probably doing. A simple wind vector analysis isn't enough, it changes based on area traffic activity as well.
---
Q3. At my airports, it gets so busy sometimes on nice weekends that you can hardly get a word in edgewise on the CTAF. This would only make it worse. EDIT: The video said it announces wind shear (how would it know without multiple sensors around the airfield) and temperature/dewpoint when needed. *I like that fact that that the AI can eliminate viz and temp/dewpoint when not relevant.
REPLY - We developed this in Potomac Airfield's pre-911 insanely-busy CTAF, then topping over 70,000 operations per year in our tiny airspace. It was, and sometimes still is, a cattle-auction on 122.8. Got it. Absolutely.
This was one of the most critical frequency sharing challenges to overcome.
The 'art' of sharing super-busy unicom/CTAF is not easy to do, both from a technical hardware standpoint as well as smart software. We have polished this at over 150 locations worldwide so far, at many super intense locations.
The objective achieved is it behaves 'as would an experienced pilot to another pilot, on a super busy VHF.
Our proprietary algorithms for windshear were best summarized by Bill F, then FAA's program manager for LLWAS, who was based at Potomac. He said to me, "This thing works better than the LLWAS at DCA, how does it work?"
It's proprietary but works really really well.
I explained it to a gov contractor under NDA one time. He said, "That's really clever. Dont tell anyone, we sell LLWAS for over $1m each"
---
Q4. Question: Is it on the ATIS frequency or the CTAF freq? EDIT: It broadcasts on the CTAF frequency. Adds to freq congestion. Not good on a busy Saturday afternoon. Why not put this system on the weather freq?
REPLY - For acceptance by FAA and FCC we had to prove that we did NOT interfere.
Which we did.
Imagine that, proving a negative to TWO federal agencies!
As to why not a 'weather frequency,' a couple of very good reasons.
Continuously able to monitor busy unicom airport traffic - Coming into Potomac pre 911, it was not unusual to have 5 or more aircraft in our tight pattern inside the CLass B for DCA and AFD. The last thing you'd want to do coming into this hornet's nest is to go OFF the unicom (CTAF) to listen to a 2 minute or more transmission from miles away. Staying on Unicom is/was he only way to reduce the probability of a mid air.
'Weather frequencies' not available - Also, obtaining the required set aside air traffic VHF is non-trivial. Under ideal conditions the process with FAA and FCC takes 2-3 years. I know, I did it and coordinated several times. In many metropolitan areas the limited number of available air traffic frequencies have ALL been reserved for future use by area ATC facilities, so there are no frequencies available.
So continually monitoring the traffic frequency, and avoiding the 2-3 year procees, if even possible, are HUGE benefits, as long as the automation has proven it does not interefere.
---
Q5. Could you tie into a ground based ADSB system and announce other traffic in the area/pattern? Now THAT would be slick!
REPLY - Great minds think alike! Yes, we already have had patents on this, and more coming. And yes, we already have developed a multi-spectrum sensor that will see 'everything.' 9Not JUST adsB)
Again, I don't want to publish what we are doing, but available to chat offline.
---
Q6 - Sell me on it if I manage three high-end private airfields. Why should we spend the money when we need so many other things???
REPLY - It is literally "The ultimate airport improvement."
Not exaggerating, pilots at our airports love these systems.
It gives any airport the ability to be what was once called 'attended' 24/7.
It is the MOST 'visible' user-enhancement possible, short of a staffed control tower 24/7.
My wife, a pilot came up with the great tag line, "Imagine your airport having tower like services, 24/7, for just $$$"
--
Beyond having a runway, and lighting, maybe a PAPI or two, and maybe an IFR approach, the next step would normally be an AWOS.
While everyone would like an AWOS, have you checked into their up-front or ongoing costs?
And AWOS only give METAR, no other warnings, no radio check, etc. No scanning and reporting ELT's. Just METAR.
Why doesn't every airport have a regular AWOS? - The only airports in USA with AWOS are those where after enough paperwork FAA funds will eventually subsidize the $300K+ up-front project cost, and that are then willing to carry the $5-10-15k per year to keep an AWOS functioning.
Check around. Update me if you find any different.
Many airports that are eligible choose not to accept FAA grants, because forever after everything they have to do must be FAA pre-approved, done with the same complexity and high cost as FAA would do it. For example, if I accepted $1 from FAA, my required 'grant assurances' for that $1 would prevent me from using the non-approved LED runway lights that save money, never burn out, and my pilots love. Even my ability to rent hangars at market prices would be dictated by FAA policies. So lots of airports that could accept FAA funds, dont.
Again, that all said, our pilots, or pilots that have flown into our airports, love these systems.
Q7. EDIT: I watched more videos, but I still didn't see traffic advisories? If the system does this, is there a video showing such in action?
REPLY - The system does not understand the specific words being spoken, but there is a great deal that can be derived from patterns of communication.
I'd be delighted to chat about it offline.
For most of my life I have often had to restrain my enthusiasm to not explain everything I do to everyone. The inventor / enthusiast / philosopher wants to, but the patent-holder has (painfully) learned to refrain from doing so.
Q8 - Suggestion: 1. Make the advisory stop broadcasting if an incoming broadcast is sensed??? Its an idea to reduce req congestions, which is an issue for me 2. Instead of the ctaf, put it on the weather freq at busy uncontrolled fields
See above. One of the key technical challenges (solutions) is having the ability to detect the transmission of others BEFORE they begin, so as to not step on them when they do.
As to 'weather frequency,' see above. The process to obtain use for a discrete frequency has many many many barriers. The 'unofficial use' of a frequency (wink wink) exposes the airport to being shut down by FCC and fines. For the record, we have never used this to beat anyone up, but they seem to eventually draw fire all on their own.
In theory, in the middle of nowhere, at very low power that would not be detected very far, one could operate without drawing bureaucratic fire immediately. But even then, for example, one nice chap on the Eastern shore bought a low-power knock-off for his ranch airstrip, soon to discover it was interfering with busy airports nearby.
Even if you just stick it on a presently 'un-occupied' frequency, 96% chance that frequency has already been reserved by some ATC facility for future use, and you get stuck in the goo.
Send a text to me at 202 575 5700 and we can chat?