INS alignment

[grunf]

How is the INS alignment implemented? These things can take quiet some time (10-15 min) to complete? :)

[drPhibes]

IIRC, the Viggen doesn't have a traditional INS, but a navigation system based on the air data computer. But I have been wrong before (once), so I'll have to check my manuals after work.

[RaXha]

Edit: Better explanation here

Edited November 29, 2016 by RaXha

[grunf]

Thank you guys!

[microvax]

Our variant luckily has radar altitude navigation for fixes. Accuracy is better then 100m afaik.

[grunf]

I really hope we'll get the manual before release. Reading the JA37 one can lead to wrong assumptions. :D

[renhanxue]

The AJ and AJS 37 only have a "pseudo-inertial" system. The system is based on dead reckoning, that is it continuously calculates the aircraft's position based only on time elapsed and airspeed (while correcting for wind speed, either forecast wind as pre-flighted into the computer or wind measured by comparing airspeed to speed over ground as measured by the radar altimeter system).

 

The upside of this is that you never need to wait for INS to align before takeoff. The downside is that it drifts quite significantly. It's not so bad at all if you're flying in a straight line with the radar altimeter on, but if you start maneuvering it can get sorta confused pretty quickly. To compensate for this you occasionally need to take "fixes", that is update your known position by the use of some recognizable terrain feature. You can either take a radar fix (measure the distance and angle to the feature in question) or an optical fix (press a button when you fly over it).

 

On the AJS 37 they added a height map database that will try to correct your position automatically by fitting your estimated position and the radar altimeter measured altitude against the height map, but the devs mentioned in the reddit AMA that this of course only works if you're flying over terrain with some recognizable height differences...

[Airbusdriver]

Well acuraccy and drift should't be that bad. In the book system 37 there is a interesting story on that.

During the first test flight of the navigation system the 60s, they took of and flew about 10 mins towards the first waypoint which was a church. On arriving over the church the computer calculated the waipoints location still within the graveyards outside the church so drift was clearly within accceptable limits. This was done only with forcasted winds entered before takeoff. The implementation of the navigation doppler radar improved the system further by correcting the windinformation to the computer.

And then the ternav introduction with the AJS update improved the system further by using the radioaltimeter terrain profile against a database.

So probaby only expect to do 1-3 position fixes per mission. Either by radar or optical.

Its performed by chosing your normal navigation waipoints on clearly visible terrain objects, like the tip of an island or a distinct corner of a lake. Then you will see the differens of the actual position and the calculated waipoint position on the radar screen. Press the fixbutton in the first detent then move the radar cursor to the correct location and depress the button fully. Then the system adds that position difference so you navsystem is updated. If the radar is off, look at your map and follow your line on the map visually, when you overfly the waipoint press the same button all the way and you have now done a optical position update.

If your waipoint is defined as a target waipoint then any update will not correct your postion but instead correct the target waypoint position. This is also done automatically when firing so if you are coming back for a second attack run you have a very good position of your target.

Clerver system! ;-)

[pewbeef]

The AJ and AJS 37 only have a "pseudo-inertial" system. The system is based on dead reckoning, that is it continuously calculates the aircraft's position based only on time elapsed and airspeed (while correcting for wind speed, either forecast wind as pre-flighted into the computer or wind measured by comparing airspeed to speed over ground as measured by the radar altimeter system).

 

The upside of this is that you never need to wait for INS to align before takeoff. The downside is that it drifts quite significantly. It's not so bad at all if you're flying in a straight line with the radar altimeter on, but if you start maneuvering it can get sorta confused pretty quickly. To compensate for this you occasionally need to take "fixes", that is update your known position by the use of some recognizable terrain feature. You can either take a radar fix (measure the distance and angle to the feature in question) or an optical fix (press a button when you fly over it).

 

On the AJS 37 they added a height map database that will try to correct your position automatically by fitting your estimated position and the radar altimeter measured altitude against the height map, but the devs mentioned in the reddit AMA that this of course only works if you're flying over terrain with some recognizable height differences...

 

So this differs from traditional INS in that you do not utilize gyros and accelerometers?

 

But I assume one would still have to take heading (compass) and preferably also pitch into account together with the airspeed and time elapsed?

[drPhibes]

It uses a combination of the air data system (which measures total pressure, static pressure, air temperature and angle of attack), a 3-axis accelerometer, the HSI and ADI. All this info is then wind compensated (either via doppler radar or forecast values), and the aircraft's ground speed in X, Y and Z axis is calculated every 103ms. By numerically integrating speed you get the distance traveled, which allows you (or the CK37 to be precise) to keep track of your current position.

[pewbeef]

It uses a combination of the air data system (which measures total pressure, static pressure, air temperature and angle of attack), a 3-axis accelerometer, the HSI and ADI. All this info is then wind compensated (either via doppler radar or forecast values), and the aircraft's ground speed in X, Y and Z axis is calculated every 103ms. By numerically integrating speed you get the distance traveled, which allows you (or the CK37 to be precise) to keep track of your current position.

 

Ok, so it still uses accelerometers? Why not integrate the acceleration and then integrate the speed obtained from that integration? This would eliminate the air data errors.

 

And what is the defining difference from a traditional INS?

 

I'm sorry if I'm a bit slow and have missed something obvious here, just curious as to how the system works and why the method of navigation was chosen over traditional INS. :book:

[RaXha]

Ok, so it still uses accelerometers? Why not integrate the acceleration and then integrate the speed obtained from that integration? This would eliminate the air data errors.

 

And what is the defining difference from a traditional INS?

 

I'm sorry if I'm a bit slow and have missed something obvious here, just curious as to how the system works and why the method of navigation was chosen over traditional INS. :book:

 

The biggest difference is the lack of a gyros that needs several minutes to align before moving the aircraft (like the one in the A-10C, and that has a very fast modern INS system compared to what was available in the 60s). Since the Viggen was designed to scramble fast to stop an oncoming invasion a traditional INS might not have been optimal for the task. With a skilled pilot i believe i read that the startup time for a Viggen could be as low as 1-2 minutes.

 

So this differs from traditional INS in that you do not utilize gyros and accelerometers?

 

But I assume one would still have to take heading (compass) and preferably also pitch into account together with the airspeed and time elapsed?

 

Yes, i believe the pilot will still need to check heading every now and then, runway heading is entered just before rolling on takeoff for example to make sure the computer calculates the traveled route correctly.

Edited December 2, 2016 by RaXha

[renhanxue]

Because this is still a 60's aircraft at its heart. Getting accelerometers precise enough for navigation purposes was very complex, expensive and above all bulky back then, and the Air Force did not at all like the readiness penalties imposed by having to sit around and wait for half an hour and wait for the gyros to stabilize either. 60s INS alignment really could take that long.

[pewbeef]

The biggest difference is the lack of a gyros that needs several minutes to align before moving the aircraft (like the one in the A-10C, and that has a very fast modern INS system compared to what was available in the 60s). Since the Viggen was designed to scramble fast to stop an oncoming invasion a traditional INS might not have been optimal for the task. With a skilled pilot i believe i read that the startup time for a Viggen could be as low as 1-2 minutes.

 

Yes, i believe the pilot will still need to check heading every now and then, runway heading is entered just before rolling on takeoff for example to make sure the computer calculates the traveled route correctly.

 

I see, gyros is the deal. And the scramble tactics makes sense for going without gyros.

 

Because this is still a 60's aircraft at its heart. Getting accelerometers precise enough for navigation purposes was very complex, expensive and above all bulky back then, and the Air Force did not at all like the readiness penalties imposed by having to sit around and wait for half an hour and wait for the gyros to stabilize either. 60s INS alignment really could take that long.

 

Also makes sense. It is all a trade off between functionality and penalties.

 

Thank you for your answers!

[Pocket]

Since the Viggen was designed to scramble fast to stop an oncoming invasion a traditional INS might not have been optimal for the task. With a skilled pilot i believe i read that the startup time for a Viggen could be as low as 1-2 minutes.

.

 

When we were stationed on Gotland during the fall of the USSR the scramble time was less than two minutes if the readiness state of the flight was at the second highest level (systems on, engine not running). At the highest level, the engine was kept running.