Forum: F-35 Avionics

Can the F-35 Avionics + Sensor package be moved to the F-22?



Search Search  Register Register  Private Messages Private Messages
guidelines Forum Guidelines
Post new topic   Reply to topic   1, 2  Next
View previous topic Log in to check your private messages View next topic
Author Message
kamenriderblade
PostPosted: May 15, 2013 - 05:11 AM Reply with quote Back to top
Senior member
Senior member


Joined: Nov 24, 2012 - 02:20 AM
Posts: 435
Location: USA
Status: Offline
Some time long after the F-35 becomes IOC, will the F-22 be able to upgraded to have equivalent sensors + avionics to the F-35?
 View user's profile Send private message  
 
Sponsor
New postPosted: May 19, 2013 - 5:26 PM Back to top
F-16.net Sponsor





  Send private message  
 
popcorn
PostPosted: May 15, 2013 - 05:48 AM Reply with quote Back to top
Elite 2K
Elite 2K


Joined: Sep 24, 2008 - 09:55 AM
Posts: 2028

Status: Offline
kamenriderblade wrote:
Some time long after the F-35 becomes IOC, will the F-22 be able to upgraded to have equivalent sensors + avionics to the F-35?


Sensor fusion and MADL to create a sharable COP leverages the strengths of both plarforms.
 View user's profile Send private message  
 
jacarlsen
PostPosted: May 15, 2013 - 06:34 AM Reply with quote Back to top
Active Member
Active Member


Joined: Jul 02, 2004 - 01:08 PM
Posts: 102
Location: AIM Norway, Kjeller, Norway
Status: Offline
According to wikipedia the apg-77v1 draws heavily on the apg-81.
 View user's profile Send private message  
 
hb_pencil
PostPosted: May 15, 2013 - 11:21 AM Reply with quote Back to top
Forum Veteran
Forum Veteran


Joined: Aug 18, 2011 - 10:50 PM
Posts: 540

Status: Offline
kamenriderblade wrote:
Some time long after the F-35 becomes IOC, will the F-22 be able to upgraded to have equivalent sensors + avionics to the F-35?


This was the plan back about 2008ish. It was cancelled because of excessive cost. This is the key problem with the fighter: the avionics suite has been found to be extremely difficult to upgrade. Various attempts to do so have been marred by excessive costs. We shall see what the AF does in the future.
 View user's profile Send private message  
 
popcorn
PostPosted: May 15, 2013 - 01:38 PM Reply with quote Back to top
Elite 2K
Elite 2K


Joined: Sep 24, 2008 - 09:55 AM
Posts: 2028

Status: Offline
hb_pencil wrote:
kamenriderblade wrote:
Some time long after the F-35 becomes IOC, will the F-22 be able to upgraded to have equivalent sensors + avionics to the F-35?


This was the plan back about 2008ish. It was cancelled because of excessive cost. This is the key problem with the fighter: the avionics suite has been found to be extremely difficult to upgrade. Various attempts to do so have been marred by excessive costs. We shall see what the AF does in the future.


Sounds like something Dr.,Frankenstein might attempt. AFAIK, no new developments in this direction and with the Sequester, even more unlikely to happen.

http://www.defensenews.com/article/2011 ... -New-Brain
 View user's profile Send private message  
 
popcorn
PostPosted: May 15, 2013 - 02:18 PM Reply with quote Back to top
Elite 2K
Elite 2K


Joined: Sep 24, 2008 - 09:55 AM
Posts: 2028

Status: Offline
hb_pencil wrote:
kamenriderblade wrote:
Some time long after the F-35 becomes IOC, will the F-22 be able to upgraded to have equivalent sensors + avionics to the F-35?


This was the plan back about 2008ish. It was cancelled because of excessive cost. This is the key problem with the fighter: the avionics suite has been found to be extremely difficult to upgrade. Various attempts to do so have been marred by excessive costs. We shall see what the AF does in the future.


Sounds like something Dr.,Frankenstein might attempt. AFAIK, no new developments in this direction and with the Sequester, even more unlikely to happen.

http://www.defensenews.com/article/2011 ... -New-Brain
 View user's profile Send private message  
 
lamoey
PostPosted: May 15, 2013 - 03:34 PM Reply with quote Back to top
Forum Veteran
Forum Veteran


Joined: Apr 25, 2004 - 06:44 PM
Posts: 699
Location: 77006
Status: Offline
IIRC the F-35 uses an industry standard programing language, but the F-22 does not. This makes it excessively expensive to put the F-35 systems on the F-22 as most of the software on the F-22 would need redoing.

_________________
Former Flight Control Technican - We keep'em flying
 View user's profile Send private message  
 
kamenriderblade
PostPosted: May 15, 2013 - 07:14 PM Reply with quote Back to top
Senior member
Senior member


Joined: Nov 24, 2012 - 02:20 AM
Posts: 435
Location: USA
Status: Offline
If that was the case, who was the fool who made the decision to have the Raptor not use an industry standard programming language?
 View user's profile Send private message  
 
SpudmanWP
PostPosted: May 15, 2013 - 07:56 PM Reply with quote Back to top
Elite 3K
Elite 3K


Joined: Oct 12, 2006 - 08:18 PM
Posts: 4266
Location: California
Status: Offline
It’s not that simple. The F-22’s software is written in ADA which is a vast improvement over what avionics used to be done in.

Quote:
Ada's Ancestry
Back in 1957, a computer scientist at IBM Corporation named John Backus created a new language that was intended to make it so that scientists, engineers and mathematicians could more easily solve mathematical and scientific problems. The language, FORTRAN, which is short for "FORmula TRANslation," was a complete success; forty years later, it's still being actively used and developed.

Across the Atlantic, in Europe, some computer scientists who had known about FORTRAN formed a committee to create their own language, one that they could use for the same purposes as FORTRAN. Only months later, in 1958, the new programming language, dubbed "ALGOL," an acronym for "ALGOrithmic Language," was all finished, and quickly standardized.

It's often considered "tradition" to name a computer language standard after the year that it was standardized, so the creators of ALGOL named it ALGOL 58. People all over Europe continued to work on ALGOL, and only two years later, ALGOL 60 was unleashed.

ALGOL 60 is very important in the history of several languages used today, including Ada. From ALGOL 60 we get three language families:

SIMULA, created by a group of computer scientists at the University of Oslo in Norway.

CPL, created by Ken Thompson of AT&T Bell Laboratories for the purpose of writing the MULTICS operating system, which was a precursor to Unix. CPL is the ancestor of C.

Last but not least, there was Pascal, written by Niklaus Wirth of the Swiss Federal Institute of Technology in Zurich, Switzerland for the purpose of teaching programming to college students, and it is from Pascal that we get Ada.


The Contest
In the 1960s and 1970s, the United States Department of Defense was using more than 2,000 languages for its mission-critical programming. Most of these were languages that were developed for one specific job. Finally, in 1975, the DoD formed the U.S. Department of Defense High-Order Language Working Group (HOLWG) to find a solution to what was often called the "software crisis."

What the HOLWG group members decided was that they needed to create a language that they could use for just about anything, whether it be systems programming, artificial intelligence, and, most important of all, real-time programming and embedded systems. Real-time programs are the programs used for controlling such things as traffic lights, guided missiles, and bar-code scanners. Embedded systems are the small computers that are built into most modern cars, airplanes, and stereos.

Rather than create this new language themselves, they decided to hold a contest. Several teams joined, each represented by a color. Coincidentally, all of the teams created Pascal-based languages. In the end, the winner was the green team -- CII Honeywell-Bull in France. Eventually, the language was christened "Ada," in honor of Lady Ada Lovelace, daughter of famed poet Lord Byron and assistant to mathematician Charles Babbage, who invented the Analytical Machine. Lady Ada is often considered to be the world's first programmer.

In 1979, the DoD created its first draft documentation on Ada, and the language was first standardized in 1983. Now named "Ada 83", this standard was originally controlled entirely by the DoD, and nobody outside the DoD could create any Ada compiler without the authorization of the Defense Department.

All that changed in 1987, however, when the DoD released Ada to the public and the language was made an international standard by the International Standards Organization (ISO). By 1990, over 200 validated Ada compilers had been produced, and in 1995 a new standard, called Ada 95, was announced. Ada 95 is object-oriented, and offers interfaces to the languages C, FORTRAN and COBOL.


ADA in the F-22
Ninety percent of the software for the F-22 is written in Ada. Exceptions to the Ada requirement are granted only for special processing or maintenance requirements.

For some of the higher risk software, such as sensor data fusion, specific algorithm testbeds have been constructed. Prototype software, which is instrumented to measure performance (correlation times, accuracy, etc.) has been operational since the start of Engineering and Manufacturing Development phase of the F-22 program.


More info on ADA here http://en.wikipedia.org/wiki/Ada_%28pro ... anguage%29

There were many lessons learned when LM went to the F-35:
1. They whet with C++ instead of ADA to broaden the programmer base and to use standard development tools
2. They use “virtual machines” (VMs) extensively (no core specific functions)
3. They use middleware extensively so that a change in hardware does not require a complete rewrite and recertification of the avionics suite
4. All major Avionics components and weapon stores are connected via fiber-optic interconnects which allows for future upgrades without rewiring the jet.
5. The F-35 isolates, in discreet VMs, the individual processing of avionics data then fuses the results. The F-22 tended to run multiple functions in a single computer instance so if there was a problem, it affected multiple functions (remember the dateline issue).
6. The F-35 was planned for all IOC functions in its current hardware state and only every other Block requires significant hardware changes (likely adding/upgrading ICP cards).

Unfortunately not a lot is known about the F-22’s architecture to draw any more detailed explanations besides these.

Btw, The FY2014 budget shows that the Hardware/Software/Labor (does not include dev) costs of the F-22 3.1 upgrade is about $1.5mil per plane. This is the same as the structural changes that are also scheduled for the F-22. Here is what hardware it includes.

Quote:
Increment 3.1 hardware modifications enable the F-22 to host Increment 3.1 capabilities including retargeting Joint Direct Attack Munitions (JDAM), carriage and delivery of the Small Diameter Bomb (SDB) on preplanned missions using the Joint Mission Planning System (JMPS), using Synthetic Aperture Radar (SAR) Air-to-Ground radar mode to permit attack of emerging targets, and saving SAR imagery onto the Tactical Data Transfer System (TDTS) for post-mission analysis.

Increment 3.1 modernizes F-22 aircraft with Enhanced Central Integrated Processor (ECIP) cards, a new Map Inter-Face (MIF) board for the 4th generation APG-77VI radar, a modification to the GPS Inertial Navigation System (GINS), and a TDTS.

Installation of these kits is done either at the depot or by a contractor field team depending on the complexity of the effort.

Increment 3.1 also includes Mission Support Requirements.

Studies were done in FY2012 and lab development in FY2013 to move the F-22 into more of a Open Architecture environment. Results have not been released yet.

Budget data on the F-22’s 3.2 upgrade will not be available till the FY2016 timeframe.

_________________
"The early bird gets the worm but the second mouse gets the cheese."
 View user's profile Send private message  
 
fiskerwad
PostPosted: May 15, 2013 - 10:02 PM Reply with quote Back to top
Forum Veteran
Forum Veteran


Joined: Nov 13, 2004 - 07:43 PM
Posts: 706
Location: 76101
The mandate by DOD to use ADA was a disaster. The language was designed by committee, the certified compilers were being developed as the language changed, NO real time code was ever determined to actually operate in real time, and the number of qualified ADA coders was only ever just a handful.
The biggest advantage for software today is the speed and capability of the hardware available. Memory size and speed, processor speed, and access times are what made today's avionics what they are.
fisk

_________________
Mipple?
 View user's profile Send private message  
 
munny
PostPosted: May 16, 2013 - 12:47 AM Reply with quote Back to top
Forum Veteran
Forum Veteran


Joined: Jan 13, 2010 - 01:39 AM
Posts: 527

Status: Offline
fiskerwad wrote:

The biggest advantage for software today is the speed and capability of the hardware available. Memory size and speed, processor speed, and access times are what made today's avionics what they are.
fisk


Also security through obscurity.
 View user's profile Send private message  
 
hb_pencil
PostPosted: May 16, 2013 - 04:14 AM Reply with quote Back to top
Forum Veteran
Forum Veteran


Joined: Aug 18, 2011 - 10:50 PM
Posts: 540

Status: Offline
fiskerwad wrote:
The mandate by DOD to use ADA was a disaster. The language was designed by committee, the certified compilers were being developed as the language changed, NO real time code was ever determined to actually operate in real time, and the number of qualified ADA coders was only ever just a handful.
The biggest advantage for software today is the speed and capability of the hardware available. Memory size and speed, processor speed, and access times are what made today's avionics what they are.
fisk



I don't know if it was ADA's fault. The language has widespread use because it is considered extremely solid. Very easy to test and has extreme reliability. Its by no means an "obscure" language either. Most Commercial aircraft use it for safety critical functions (I believe), as do Nuclear power plants, ATC systems, Trains ect.

Problem is that its not well developed outside of that. Libraries are just not there for a lot of things, requiring significant independent development to create (which then is a problem for future development as these are basically bespoken code). Situation has changed for the better since the 1990s and there is more tools libraries ect to undertake development with.

In reality I think the program suffered from multiple problems, which ADA was one of. Changing requirements, bad architecture design and some of the problems you note combined to make the mess you see today. Spudman's list is really good, as it shows that its not just the languages' fault.

I wish Neurotech would weigh in here. He probably has some good insights.
 View user's profile Send private message  
 
neurotech
PostPosted: May 16, 2013 - 06:50 AM Reply with quote Back to top
Elite 1K
Elite 1K


Joined: May 09, 2012 - 10:34 PM
Posts: 1256

Status: Offline
Your correct on the first part. ADA is extremely reliable when written according to well established guidelines. The claim about "a handful of ADA developers" is misleading. There might be many thousands of C/C++ developers out there, but very few who could write DO-178B Level A or MIL-STD-478 code. I know almost none of the C/C++ code I wrote for simulation and testing made it into the fleet aircraft. Some of the mission systems on the F/A-18E/F are written in C/C++, but those systems are not safety critical like the FBW.

The reason the F-22 is such a Charlie Foxtrot when it comes to software and avionics, is that they held onto legacy technology for too long. By the time the jet went IOC, They could practically find spare parts in a museum. The Intel i960 processors used in ground equipment are actually re- purposed from old laser printers or drive controllers. The other thing is that architecture (i960 RISC) is not very common, and the only other aircraft that uses the i960 is the HAL Teja which is also a 1980s design that went IOC much later.

Some of the newer F-22 CIP boards have more powerful PowerPC chips, along with FPGAs (incorrectly referred to as ASICs in some literature) that have more compatibility with the F-35 and can share parts of the code base. It would be highly impractical to do a "brain transplant" of the entire F-35 avionics suite. Various components (EW recievers etc) are simply not compatible, and the radar is a generation behind the F-35, even though they share common development heritage. The result would be a mix of technology that would still be a major development program, not a low-cost move.

If Lockheed were to transplant the F-35 avionics and mission systems into a F-16V Block 70, it would cost almost as much as a F-35, the cost savings wouldn't be there. The lower rate of production of a F-16V would offset any savings on the airframe cost, compared to the F-35.

As for the library availability is only minor factor, especially with 80s technology. The ability to use off-the-shelf code packages in avionics is the difference between night and day, and is more a about development methodological popular in the 90s, than being a language issue. In the 80s, Some function blocks could have been borrowed from other projects, but major amounts are written for each project. Development tools are more than just libraries, and the validation and debugging of the new code has improved considerably. Logical programming errors can still cause problems, such as confusing metric and imperial units.

No matter how much processing power, speed and memory size is available, bugs will cause problems, and newer avionics might be smaller and lighter, but its not automatically bug-free.

I'm not sure precisely what architecture SpaceX Falcon 9/Dragon uses, but its a combination of VXWorks (C/C++) RTOS and Linux, which is pretty much COTS software for avionics. Elon Musk talks about test profiles where they "fly" a mission by using simulated inputs into a real flight computer, its not a software simulated mission computer. Running the flight computer through functional tests is really the only way to know for sure things are working, and even then problems still occur on the actual flights.

On the F/A-18E/F we would run through exercises in the simulator, designed to test new software and mission systems. Doesn't matter how many dev tools, how many code scans are available, it all has to be tested in realistic conditions. It's amazing at how much a FPGA "signal generator" can replicate flight inputs on the ground. This data was obtained from actual flight tests.

In any safety or mission critical system, changes have to be fully tested, and making changes mid-course gets expensive. It takes real flight time and many man hours.

The F-22 avionics and mission software gets somewhat unfairly criticized because the jets are so damn expensive, and a limited number of aircraft. The EA-18G program had just as many unique requirements, as the F-22, but on paper each jet isn't costing $400m Program Unit Cost. In reality, the physical unit cost of a EA-18G with ALQ-99s fitted and loaded, isn't much less than a F-22. The AMCs are several generations newer than the F-22, and has high commonality with the F/A-18, so people don't complain as much about development costs.
 View user's profile Send private message  
 
kamenriderblade
PostPosted: May 17, 2013 - 05:53 AM Reply with quote Back to top
Senior member
Senior member


Joined: Nov 24, 2012 - 02:20 AM
Posts: 435
Location: USA
Status: Offline
I can see the future of all aircraft Avionics software being written in the F-35 overall style

C/C++ as the base language for all software aspects.

Everything running on isolated VM's.

Middleware to isolate hardware from software so easy switching of hardware won't cause headaches in software.

It's a superior architecture IMO, using the latest COTS Hardware/Software practices.

Moving forward, I can see all aircraft across the entire aerospace industry going that way with standardized avionics software design.
 View user's profile Send private message  
 
neurotech
PostPosted: May 17, 2013 - 06:25 AM Reply with quote Back to top
Elite 1K
Elite 1K


Joined: May 09, 2012 - 10:34 PM
Posts: 1256

Status: Offline
kamenriderblade wrote:
I can see the future of all aircraft Avionics software being written in the F-35 overall style

C/C++ as the base language for all software aspects.

Everything running on isolated VM's.

Middleware to isolate hardware from software so easy switching of hardware won't cause headaches in software.

It's a superior architecture IMO, using the latest COTS Hardware/Software practices.

Moving forward, I can see all aircraft across the entire aerospace industry going that way with standardized avionics software design.

Despite what people claim, it isn't isolated VMs at the lower level for hard real-time like FBW. At the higher level, yes, it is isolated processes in a hypervisor VM.

As for the rest of it, its easy to say "moving forward" but in reality, there is a lot of legacy code in jets like the F-15 & F-16, and yes, F-22, that makes porting legacy ADA code into C/C++, basically a rewrite, such a lengthy process.

Unlike some of the training or "full mission" simulators, when doing technical evaluation, we sometimes used actual components off actual F/A-18s. We could literally pull a AMC-3 out of a jet, and put it in the simulator unit, and fly the mission, evaluating the software function.
 View user's profile Send private message  
 
Display posts from previous:     
Jump to:  
All times are GMT + 1 Hour
Post new topic   Reply to topic
View previous topic Log in to check your private messages View next topic