Instructables

Hacking the Failed Billion Dollar Border Security System

As much as I hate the phrase, it is a major FAIL!!!

          "After revealing last week that a pilot installation of controversial, buggy border-security
          scanner towers had finally been accepted into service, the US government has now
          admitted that the project is a technical failure..."
 
          - Boeing’s Border Watchtowers Can’t See Straight, Wired Magazine


I worked on the project two years ago, very briefly.  I knew it was a basket case the moment I saw it.  Not only do I know exactly how it happened, and why it ended up like it did, I can explain the whole thing in three words... North American Union.

This instructable will show you how to fix the failed $1B border security system in three easy steps.  The goal is to create a system using only what is already there, that is fully autonomous, and which operates as good as, or better than, a system relying on having one or more human operators present at all times.
 
Remove these adsRemove these ads by Signing Up

Step 1: Fix the Filter

Picture of Fix the Filter
kalman.png

Two years ago I got hired as a consultant by Boeing with the job description of making sure everything worked.  On my first day on the job, I read some documentation that said the system used "Kalman filtering."  It also said it used "particle physics?"  Well, I don't know much about particle physics, but I do know my Kalman filtering.  Later  that day I asked my manager if he could tell me how the Kalman filter was implemented.  He told me that I should only concern myself with "intelligent" questions.  I knew right then and there that the system would be a complete and utter failure.  I was let go after 30 days for "incompetence."

One of the biggest complaints about the Boeing system is that it's constantly giving false alarms. Well, one thing a Kalman filter is good for is eliminating false alarms.  A Kalman filter is a software algorithm that is used for all kinds of things. Most often it's used for guidance and navigation, but it can be used for everything from economic projections to population growth studies.

The Kalman filter is my favorite filter in the world.  If you walked up to me in a bar and asked what my favorite filter was, I'd say the Kalman, by far.  One reason it's my favorite is because, if you read the description on Wikipedia, it looks like a big ugly hairball of mathematical equations!  Go here and scroll down (actually, look at the screen shot above.)  Beautiful, ain't it!  BTW, that's one problem with Wikipedia.  It sometimes ends up being a giant sandbox for grad students working on PhD theses, which, in my humble opinion, is where that description belongs, not on Wikipedia.  The average reader isn't going to be able to understand a description like that.

Fortunately, like most complicated things, if you boil it down, it's just a tiny little common sense equation.  It is basically just a weighted filter.  For example, say you have a pile of watermelons with a sign saying "10 lb. watermelons."  Now, to estimate how much each watermelon actually weighs, you start by saying your initial estimate of the weight is 10 lbs, because that's the only information you have.  This is called the prediction.  Then you pick up a watermelon from the pile and put it on the scale and it reads 9 lbs.  But the scale says it's accurate to plus or minus 5%, meaning 95% accurate. Therefore you can only say for sure that the watermelon weighs at least 9 * .95 = 8.55 lbs.  To account for the other 5%, you take the previous estimate, your initial estimate, times 5%, and add that to the intermediate estimate, which yields 10 lbs * .05 = 0.5. Then, your overall new estimate, called the update, is 8.55 + 0.5 = 9.05 pounds. And you just keep doing that for each watermelon, resulting in a recursively filtered estimate. Regarding the covariance matrix, that is for things that have a small side effect on the system, such as if very high humidity causes the watermelons to be a tiny bit heavier, in which case you would add humidity to the covariance matrix.

With the Boeing system, a specific complaint was that, on windy days, there would be even more false alarms due to things like tree branches swaying in the wind, not to mention tumbleweeds rolling by.  To correct for this, one item for the covariance matrix would be current wind conditions.  Maybe even use "tumbleweed recognition," which, by the way is entire feasible considering that the way I do pattern recognition is by closing my eyes and visualizing the video data (the ones and zeros) and "seeing" what tumbleweeds look like over time as ones and zeros.  For one thing they are almost perfectly round.  And they move fast, about as fast as the wind, which I already know because it's in my Kalman filter.  Are you starting to see how the ability to inject simple common sense into the Kalman filter is what makes the Kalman filter so eloquent, and useful?  More detail on that in the next section.
BLACK_M3SA3 months ago

Have you though about working on your own sentry gun from the Project Sentry Gun? I feel your's would be the most accurate.

el_roboto_loco (author) 3 years ago
How come nobody's reading this??? Do you mean to tell me, that I write an article on how to salvage a billion dollar white elephant, payed for with taxpayer money, and only a bare handful of people even find it interesting enough to bother to read the article? GOD...WHAT A DULL, DULL WORLD I LIVE IN!!!

But, that's okay. I'm going to repurpose this article, and use for my next indestructable, "How to make a HAL 9000 Series Computer." I'll be building a HAL 9000, and using it to operate the newly hacked border security system.

("2001: A Space Odyssey" was on tonight.)
Well, when I read "Hacking the Failed Billion Dollar Security System", my first impression was that the instructable was about a way to go out and interfere with the system as it is now, which in turn gave be the impression that this was some kind of joke or at least something rather poorly made and based on speculation. I read it out of curiosity anyway, and my initial impression was of course incorrect. However, it would probably help to include the purpose of the hack in the title.
el_roboto_loco (author)  MDude1 year ago
If you read the title in the URL, it's "How to Fix..." but I changed it to "hacking" because I thought it would get more people to read it.
el_roboto_loco (author) 1 year ago
I forgot something in the article. The 3 words are North American Union!!!
Johenix2 years ago
I agree that the project was a cash gift to Boeing. They are common to the whole Military Industrial Complex.
Congress gets a kickback by buying stock in companies that are going to show a nice profit increase per share that multiplied by the P/E ratio to cause a share price rise. Or the congress member buys options on the shares, then sells the options after the price rises.

You were a real threat to their cozy little game. You might have produced something that actually worked.

My aproach would have been a 1950's capacitive alarm fence that if you even aproach it, it turns on the lights and sirens.

A filter for tumble weeds would be 1.) wind speed vs. object speed corelation. 2.) wind direction vs. object motion direction. 3.) object warmer than ambient? 4.) Multiple objects moving as a cluster? 5.) objects moving away from border?

Thanks for the intro to this type of filtering.
Nyxius2 years ago
I am a student working on the aerial robotics team @ OSU. I had all my grad students in neuroscience get hired and are therefore no longer able to compete on the team. I am trying to learn object recognition and AI systems in a pathetic attempt to fill the gap. I tried accessing your paper on the base 7 computation system. It appears to be down. I am also amazed at our similarity in technique. I read a bunch of books on object recognition and then promptly ignored it all and developed my own algorithm ( it seems to work well, with room for improvement as I get time). Basically I'm looking for simple solutions that achieve simple levels of bio-mimicry. I have achieved object clustering and complex motion tracking, but am still working on determining exact ranging. I just haven't had the time needed to finish developing it. Any feedback or tips would be much appreciated.
Chakazuluu2 years ago
If there is no communication between government agencies and assuming that all the people that are in control in these agencies are highly intelligent and highly educated people it would make since to me that they would recognize that the lack of their communication has sabotaged the project. My expectation is that they then knowing this they would then do something constructive about the situation. If they don't then they are they knowingly sabotaging their collective efforts. I am not understanding why and wherefore this happens.
pcopley3 years ago
This whole post just seems like someone complaining because they were fired. I somehow doubt the author could single-handedly develop a working South Korean border patrol AI.
el_roboto_loco (author)  pcopley2 years ago
I could do that easily. Do you have their phone number?
You think well and clearly. I believe the basic concepts for a "working ____ border patrol AI" are pretty simple, at least when once broken down into discrete chunks like this.

Waaaayyyyy back in the bronze age of computers, in the 1980's, some of us in college talked about things like this in pseudo-code. Now these are called high level languages! :)

If you can define what stimuls produces what response, that's conceptual programming. And any college grad with a degree in the field today can write the specific code in whatever langauge is popular at the time. It's no big deal to be creative, one just has to get away from the idea that things are too difficult to do. Or get away from the idea that only large teams of "smarter" people can accomplish the task. After all, aren't we all here on Instructables because we CAN learn? :)

Best of luck to you Crazy Robot!
CaseyCase3 years ago
I'd like to say that your ideas are brilliant, but I wasn't able to get past the "It's George W Bush's fault" whining. Sorry.
el_roboto_loco (author)  CaseyCase3 years ago
I de-politicized it now.
El-Roboto-Loco said:
"I de-politicized it now."

Aw shucks! Does that mean your article is now politick-ally currect? ;)
Once a paper is changed in that way it loses all common sense and power. Please tell us it is only partly sanitized, please!

el_roboto_loco (author)  DIY-Guy2 years ago
Well, it's politically palatable now. I've already been prevented from having a full time job for 5 years now. Don't want to piss them off even more.
vov353 years ago
I'm not one for border control, but this is doing programming correctly. This would make far too much sense to be employed by the powers that be.
Excellent article (IMNSHO). I suspect "Incompetence" is government contractor speak for " I have no idea what you're asking me, therefore, you must be a threat to my political hack position. You must be a naive, 'results' oriented personality and don't comprehend that the objective here is to bleed as much out of the federal budget while producing something that will take an additional 5 fold in funding to 'move towards a viable concept'. To actually produce what was contracted for only closes the contract prematurely and leaves my daughter's tuition at Princeton in needless jeapody. People like you are a danger to the system and the 'program' and will only raise the ire of the taxpayer (oh, those most bothersome taxpayers!) s well as embarass the Political Brass who actually promoted this foolishness instead of stationing the Military on the border like thousands of years of common sense experience has taught."
So NOW I KNOW WHAT "INCOMPETENCE" really means.
Thank you for the clarification. Yeah, that Kalman Filter thingy was cool too.
LosRojos3 years ago
Intelligent and well written article. Obviously would be positive steps.
lemonie3 years ago
Put a presentation together - make the thing more clear and concise.

The letter doesn't present well, you could be read as being delusional (examples of possible reactions in brackets):
You say "I realize this may sound too good to be true" (probably is then?) and "And no, I’m not a genius" (but you think you are?), and you don't cite the articles which you read recently (oh yeah, what articles?) - which are the entire basis of the argument.

L
el_roboto_loco (author)  lemonie3 years ago
I take back what I said about your reading comprehension. You do realize that the letter you read is dated last week? It's already been sent.

You already said that you had sent it, but that is not important. If they write back do update the topic though.

L
lemonie3 years ago
I’ve already done the complete redesign in my head.
This reads a lot like it's mostly in your head, rather than being on paper (electronic or otherwise)?

L
el_roboto_loco (author)  lemonie3 years ago
your level of reading comprehension is unparallelled.

I wasn't good at it in school though...

L
schkip19733 years ago
I can't wait for your next 'ible "how to fry a chip on your shoulder".
el_roboto_loco (author)  schkip19733 years ago
point taken
AndyGadget3 years ago
 
Hmmmm . . . .   I think I preferred your Flatulent Boss Detector.
el_roboto_loco (author)  AndyGadget3 years ago
Yeah, that was a good one wasn't it. Tell ya what, let me finish this billion dollar project first, and then maybe I can come up with another project like the FBD.