Perhaps, it goes without saying that the latest GTN 650 and GTN 750 are a godsend to users. The current software upgrades and hardware mean safer and easier operations. And Garmin has really outdone themselves with these futuristic avionics systems. From data entry, to ease of use, it’s everything a pilot could want and more. Fewer buttons, more control, and touch screen system are fabulous.
Can it ever be any better, it’s as if the engineers made it for pilots as if it was both designed, engineered by the very users who love it so much – We think it may become better in the future. How so you ask? Well, it turns out I was discussing this will a fellow think tanker, Troy Leclaire, recently and we talked about the issues with 1/2 second or 1-second delays from the satellites for data.
We were discussing various solutions, such as which information should be pre-loaded data, and use upgraded RAM to get at it. Here are some of Troy’s ideas to prevent bottlenecks in the system – combination of hardware and software upgrades;
1. Increase RAM
2. Make sure using the best possible bus speeds between processor and RAM
3. If using standard hard drives, switch to solid state drives/flash drives. (although flash drives have a limited read/write perhaps a combination, use flash for long term storage and solid state for short-term information, and have it possible for maintenance to easily swap out the flash drives after so many cycles.) Used but still, usable flash drives can then be used in other areas or sold off to the secondary market.
4. Determine what is necessary for the instant update and what isn’t. Terrain and such can be hard loaded since this usually isn’t going to change much. Things that do not change often (such as general weather) are not going to change too fast, although I know in some cases it can, however, you can use predictive algorithms to predict possible changes and base the display/transmissions around this so weather doesn’t have to be updated too often.
A plane flying over the ocean in good weather does not require that the weather is updated every second, so say that weather may only need to be updated every 20 times that a transmission is sent, rather than having constant updates, unless the weather meets certain guidelines in which case the transmissions can be increased. In the event that a plane is flying into bad weather, have the pilot capable of pushing a button that will increase the number of times that the weather is updated.
Tie this into a system that communicates to the satellite, so that if one pilot pushes the button, the satellite will feed the same results as often to other pilots in the same general area. Example, all planes in an area have weather updated every 30 seconds, one pilot notices bad weather and pushes his button, suddenly all planes in the area start receiving updates every 3 seconds.
5. Use a variation of the above for plane path prediction. Since usually (from what I understand) there are set routes and most planes stay on these routes, it may be possible to use predictive algorithms to determine where a plane is and where a plane isn’t ahead of time. If a tracked plane deviates from the predicted route, change how often that plane is updated on others displays. Planes closer to the receiver will be updated more regularly than planes further away.
5. Use military-grade burst transmissions from the satellites to the planes to decrease the amount of traffic being transmitted/received.
6. Augment the system from ground control radars for when planes are in holding patterns, or perhaps have them switch off from receiving via satellite to receiving via ground stations. Perhaps this can decrease the load on the satellite systems and increase the accuracy in the busiest areas.
7. Use the same type of video cards/monitors used by gamers. These have some of the best refresh rates around.
Indeed, these are all excellent solutions and concepts to keep the system robust, safe, and leading edge. Now then, in examining these ideas, and with regards to item number 4 above: Yes, all terrain, and non-changing data must be pre-loaded, with updates available, are very important, also consider things like; new wind turbine towers, radio antennas, power lines, airport frequency changes, and changes in airport services, noise abatement rules, etc.
Additionally on item number four, Troy makes an interesting observation, and it is an interesting concept. In other words, the algorithm will self-inquire for updates, as anticipated. Or the system will increase the refreshing updates as anticipated if the pilot does not. Also, the bandwidth of the system will concentrate on “need to know now” or “NKN information” first when prioritizing data feeds.
Now then, these are just a few concepts which could be integrated into the next generation of systems, although both Troy and I realize the current versions of these technologies are superior to anything else out there right now. Troy, of course, being a systems engineer type would improve these devices using these concepts if he were designing a system of this type.
Indeed, I agree, and thus, we have written this up, as these are very smart suggestions, some of which this system most likely already has now or in the works for the future versions. Nevertheless, most of these concepts would be perfect for the next generation of avionics. If you have any comments concerns or questions, feel free to shoot me an email.