[00:11.050 --> 00:13.890]  Welcome back to Space Security Challenge 2020,
[00:13.890 --> 00:15.070]  The Final Frontier.
[00:15.070 --> 00:16.210]  I'm your captain, Jordan.
[00:16.210 --> 00:18.210]  I'm actually not anybody's captain,
[00:18.210 --> 00:19.790]  but I can play one on YouTube.
[00:19.990 --> 00:23.130]  With no game changes to report on in the last hour
[00:23.130 --> 00:24.850]  and the game infrastructure cooled down
[00:24.850 --> 00:27.910]  to a less than molten state, as things are a steady state,
[00:27.910 --> 00:29.590]  we're gonna bring in some of the game organizers
[00:29.590 --> 00:33.630]  to talk a little bit more about the behind the scenes
[00:33.630 --> 00:35.290]  and the game infrastructure.
[00:35.650 --> 00:37.690]  We were gonna call this segment Between Two Jasons
[00:37.690 --> 00:39.970]  because we've got Jason Williams and Jason Latimer
[00:39.970 --> 00:42.110]  of Chromulence on the video chat,
[00:42.110 --> 00:43.590]  but I don't think y'all need to see my face.
[00:43.590 --> 00:45.490]  So, Jasons, can you hear me?
[00:47.710 --> 00:48.730]  Yeah, I can hear you.
[00:48.730 --> 00:50.170]  Excellent, all right, that was Latimer on the right
[00:50.170 --> 00:51.830]  and we got Jason Williams on the left.
[00:51.830 --> 00:53.630]  Thanks for joining me, first of all.
[00:53.810 --> 00:54.330]  I appreciate it.
[00:54.330 --> 00:56.770]  You guys have been pretty busy, I know, the last few days
[00:56.770 --> 00:59.370]  and, of course, even before that as well.
[00:59.370 --> 01:00.750]  Just first, general impressions.
[01:00.750 --> 01:03.190]  How are you feeling right now about how everything's going?
[01:06.270 --> 01:08.350]  I think everything's going really well.
[01:09.350 --> 01:10.150]  Sorry.
[01:10.150 --> 01:12.710]  All right, I'll direct into one of you specifically
[01:12.710 --> 01:14.450]  so I don't make you talk over yourself.
[01:14.450 --> 01:17.270]  Latimer, how are you happy right now with things?
[01:17.270 --> 01:17.890]  How are you feeling?
[01:17.890 --> 01:19.110]  I'm happy.
[01:19.290 --> 01:20.610]  We've had a little bit of a lull
[01:20.610 --> 01:23.150]  as teams have been working on challenge three and four,
[01:23.150 --> 01:27.390]  which are the harder challenges from the hacking perspective.
[01:27.950 --> 01:29.950]  So, we're kind of sitting back and watching
[01:29.950 --> 01:32.970]  and it's going pretty well today for sure.
[01:33.490 --> 01:36.330]  So, one of my other questions for you each,
[01:36.330 --> 01:37.990]  and we'll stick with Latimer since you're going,
[01:37.990 --> 01:40.330]  was what is your favorite challenge then?
[01:40.330 --> 01:42.570]  I already revealed, I think three was probably mine.
[01:42.570 --> 01:43.610]  Do you have a favorite,
[01:43.610 --> 01:44.810]  either one you worked on or one you didn't?
[01:44.810 --> 01:46.930]  And if it's a later one, no spoilers, of course,
[01:46.930 --> 01:50.130]  but feel free to tell us which one you like best.
[01:50.610 --> 01:54.630]  Yeah, I really liked three a lot and I was super excited.
[01:54.630 --> 01:56.510]  We gave the team something to work on overnight
[01:57.090 --> 01:58.950]  and to come back in this morning
[01:58.950 --> 02:01.250]  and have a couple of solves right away,
[02:01.250 --> 02:02.930]  that was awesome.
[02:03.290 --> 02:05.330]  Challenge four is going to be a challenge.
[02:05.330 --> 02:07.930]  So, I'm hoping that we get some teams
[02:07.930 --> 02:09.930]  that actually solve that this afternoon.
[02:09.930 --> 02:13.430]  We're going to, you know, looking at when we give a hint
[02:13.430 --> 02:16.490]  or two coming up, but yeah, I like challenge four.
[02:16.490 --> 02:17.630]  I think you guys described to me
[02:17.630 --> 02:18.450]  that four was probably,
[02:18.450 --> 02:20.670]  would you say the hardest overall of all the challenges?
[02:20.670 --> 02:21.530]  That was kind of the impression I got
[02:21.530 --> 02:23.250]  from talking to the folks that made them all.
[02:23.250 --> 02:26.310]  So, this is going to be important too.
[02:26.470 --> 02:28.290]  Certainly for the score in the game,
[02:28.670 --> 02:31.230]  a lot could happen in these last two solves.
[02:31.230 --> 02:32.950]  So, this is going to be a big moment.
[02:33.890 --> 02:35.250]  All right, so Jason Williams, same to you.
[02:35.250 --> 02:37.510]  Do you have a favorite challenge?
[02:38.590 --> 02:40.710]  Yeah, I think it's for me, it's going to be challenge four
[02:40.710 --> 02:42.830]  without trying to give too much away.
[02:42.830 --> 02:44.990]  As you know, we think this is the hardest challenge
[02:44.990 --> 02:48.030]  in the game and it exercises a lot of skills
[02:48.030 --> 02:50.530]  that the teams are going to need to use
[02:50.530 --> 02:51.970]  in order for them to solve this challenge.
[02:51.970 --> 02:54.590]  So, I'm really excited to see where they go with it.
[02:54.590 --> 02:56.830]  And hopefully in the next hour or two,
[02:56.830 --> 02:58.470]  we get a solve on this one.
[02:58.890 --> 02:59.410]  Awesome.
[02:59.410 --> 03:02.010]  Now, I think you each may have some components
[03:02.010 --> 03:04.590]  of some of these FlatSats that we can see kind of up close
[03:04.590 --> 03:05.630]  and maybe talk a little bit more.
[03:05.630 --> 03:08.570]  Jason Williams, just before we went live,
[03:08.570 --> 03:11.690]  you were showing me one of the boards that you had there.
[03:11.690 --> 03:13.310]  Do you want to go ahead and talk about that a little bit?
[03:13.990 --> 03:16.230]  Yeah, so on the FlatSat,
[03:16.230 --> 03:18.770]  we have our own hardware that we designed.
[03:19.270 --> 03:21.410]  And it's actually, I'll try to show it to you
[03:21.410 --> 03:22.990]  if you guys can see it here.
[03:23.070 --> 03:24.890]  It consists of a carrier board
[03:24.890 --> 03:27.210]  followed by two boards on top of it.
[03:27.210 --> 03:29.050]  One of them is a ZX board,
[03:29.050 --> 03:32.470]  which has on it an FPGA, an Artec 7 FPGA,
[03:32.470 --> 03:34.770]  which runs two soft core processors.
[03:34.770 --> 03:36.250]  Then the other one at the bottom here
[03:36.250 --> 03:38.210]  is the Raspberry Pi Zero.
[03:38.210 --> 03:40.410]  And this one is running the payload module.
[03:40.410 --> 03:41.490]  Yes, yep.
[03:41.490 --> 03:43.670]  And so here you can see the ZX board
[03:43.670 --> 03:46.430]  and on there is actually two soft core processor,
[03:46.430 --> 03:50.190]  an MSP430, which we use for running the radio baseband,
[03:50.190 --> 03:54.230]  and then a Leon 3 processor from Geissler,
[03:54.230 --> 03:56.450]  which Jason can talk more about
[03:56.450 --> 03:58.110]  that runs our flight software.
[03:58.170 --> 04:00.010]  Well, hold on a second though, before we go, Jason,
[04:00.010 --> 04:03.550]  that MSP430 soft core FPGA sounds a little familiar to me.
[04:03.550 --> 04:05.070]  I think I see something behind your shoulder.
[04:05.070 --> 04:07.610]  Was that related to that as well?
[04:07.950 --> 04:10.730]  Yes, if you can remember the badge service
[04:10.730 --> 04:13.010]  from 2013 DEF CON,
[04:13.010 --> 04:14.970]  it's actually the same soft core processor
[04:14.970 --> 04:17.730]  that was used for that competition.
[04:17.890 --> 04:19.770]  Yeah, Badger was the name of the challenge
[04:19.770 --> 04:21.190]  and I remember working on it
[04:21.190 --> 04:22.690]  and we kind of had an almost solution
[04:22.690 --> 04:24.710]  and then my team threw our broken solution
[04:24.710 --> 04:26.830]  and then another team fixed it and got it right.
[04:26.830 --> 04:28.390]  So we scored more points than that,
[04:28.390 --> 04:29.890]  but I worked on that during the finals.
[04:29.890 --> 04:30.850]  I remember that one.
[04:30.850 --> 04:32.210]  Yeah, that's really cool.
[04:32.210 --> 04:33.530]  A lot of the code is similar.
[04:33.750 --> 04:34.550]  That makes sense.
[04:34.550 --> 04:36.070]  Because you already had this sort of soft core
[04:36.070 --> 04:36.830]  that you needed to implement
[04:36.830 --> 04:38.610]  and it was compatible with what you do.
[04:38.610 --> 04:39.570]  So that makes sense.
[04:39.570 --> 04:39.870]  Awesome.
[04:39.870 --> 04:44.370]  Now the Leon processor then, or the payload,
[04:44.370 --> 04:45.550]  tell me a little bit more about that, Latimer.
[04:45.750 --> 04:47.150]  What's involved in that one?
[04:47.890 --> 04:53.030]  Yeah, so we used a Leon 3 soft core that Geissler developed.
[04:53.030 --> 04:54.210]  It's all open source.
[04:54.210 --> 04:55.170]  You can go out and get it.
[04:55.170 --> 04:56.810]  You can go buy an Arctic 7 FPGA
[04:56.810 --> 04:58.990]  and run your own Leon 3 on it.
[05:00.090 --> 05:02.930]  And so it's a Spark processor that was developed
[05:02.930 --> 05:05.150]  in conjunction with European Space Agency.
[05:05.150 --> 05:07.590]  So this is legit the same kind of processes
[05:07.590 --> 05:09.490]  that are utilized in space.
[05:09.810 --> 05:12.310]  So that's why we picked it for the competition.
[05:12.630 --> 05:13.650]  And then on top of that,
[05:13.650 --> 05:17.830]  we're running a NASA product called Core Flight Software.
[05:18.730 --> 05:20.210]  We picked that because
[05:20.210 --> 05:23.570]  that's a really legitimate flight software.
[05:24.310 --> 05:26.070]  NASA has been huge in the community
[05:26.070 --> 05:28.810]  as far as releasing their flight software
[05:28.810 --> 05:33.310]  that they are updating periodically.
[05:33.310 --> 05:34.670]  And there's like a number of applications
[05:34.670 --> 05:36.090]  that they have provided.
[05:36.090 --> 05:38.130]  And so we use that as a basis.
[05:38.410 --> 05:39.650]  And on top of that,
[05:39.650 --> 05:41.110]  we've developed some custom applications
[05:41.110 --> 05:42.930]  that actually interface with the hardware
[05:42.930 --> 05:44.330]  that we're using in the game.
[05:44.530 --> 05:46.850]  So that was our approach there.
[05:46.850 --> 05:48.850]  Now Latimer, how much would you say
[05:48.850 --> 05:50.990]  of like the overall system that you've got
[05:50.990 --> 05:52.950]  from the case to the boards,
[05:52.950 --> 05:53.910]  to the payload,
[05:53.910 --> 05:55.330]  all the different modules that are,
[05:55.330 --> 05:57.030]  how much did you develop?
[05:57.030 --> 05:59.130]  How much were off the shelf commercial components?
[05:59.130 --> 06:00.450]  How much were there other vendors involved?
[06:00.450 --> 06:01.510]  Like what was the sort of breakdown
[06:01.510 --> 06:03.370]  of the overall system?
[06:04.570 --> 06:06.650]  Yeah, so on the Flatsat itself,
[06:06.650 --> 06:10.490]  there's really like two major components,
[06:10.490 --> 06:12.450]  each with subcomponents.
[06:12.750 --> 06:16.070]  We had kind of a short timeframe to develop all of this.
[06:16.070 --> 06:18.050]  So when we met, you know, starting,
[06:18.050 --> 06:20.930]  we were talking more details in November of last year,
[06:20.930 --> 06:23.770]  we decided we need something quick that we can do
[06:23.770 --> 06:25.670]  to get sensors and actuators.
[06:25.670 --> 06:26.530]  So in order to do that,
[06:26.530 --> 06:28.710]  we bought an off the shelf training satellite
[06:28.710 --> 06:30.710]  that provides sun sensors.
[06:30.710 --> 06:31.910]  It has an IMU.
[06:32.070 --> 06:35.330]  It provides a gyro and magnetometers.
[06:35.710 --> 06:37.790]  It has torque rods, just like a real satellite.
[06:37.790 --> 06:39.850]  It has a reaction wheel, just like a real satellite.
[06:39.850 --> 06:42.490]  So, you know, that's one of the primary components.
[06:43.670 --> 06:47.350]  But that training satellite from a game perspective
[06:47.350 --> 06:48.690]  wasn't really interesting.
[06:48.690 --> 06:50.270]  It didn't really mimic, you know,
[06:50.270 --> 06:53.210]  the kind of processors that are running
[06:53.210 --> 06:55.830]  on like real satellites in space.
[06:55.830 --> 06:58.750]  So that's where we came up with the CNVH board
[06:58.750 --> 07:00.270]  that we developed that has, you know,
[07:00.270 --> 07:03.830]  that Arctic 7 and the Raspberry Pi.
[07:04.390 --> 07:07.330]  So those are the two kind of primary components
[07:07.330 --> 07:09.290]  on the flat sat.
[07:09.290 --> 07:12.090]  And is that where most of the sort of the game occurs,
[07:12.090 --> 07:13.810]  you know, where the challenge is people interacting
[07:13.810 --> 07:16.170]  and except is the payload module then a part of that
[07:16.170 --> 07:17.450]  or that I know we're gonna,
[07:17.450 --> 07:19.010]  maybe that's coming a little bit later,
[07:19.010 --> 07:21.950]  but like, where is that gonna get factored in?
[07:22.790 --> 07:24.150]  That's part of it.
[07:24.150 --> 07:25.050]  Okay, all right.
[07:25.210 --> 07:28.330]  Not gonna talk too much about that
[07:28.330 --> 07:30.730]  cause the team still have some things to figure out there.
[07:30.950 --> 07:32.910]  But, you know, the intent from the beginning
[07:32.910 --> 07:36.590]  was to take a picture of the moon.
[07:36.590 --> 07:38.010]  We've been telling the teams that,
[07:38.010 --> 07:40.310]  so the Pi is there really to provide an interface
[07:40.310 --> 07:42.710]  to the camera to allow them to do that.
[07:42.770 --> 07:45.630]  And we put some interesting things in there
[07:45.630 --> 07:48.250]  that the teams will soon find out about.
[07:48.250 --> 07:49.070]  Yeah, that makes sense.
[07:49.070 --> 07:50.590]  And I don't think anyone is surprised.
[07:50.590 --> 07:52.670]  They knew the end goal kind of going
[07:52.670 --> 07:54.210]  and I've definitely seen some chatter from them.
[07:54.210 --> 07:56.090]  So that shouldn't be too much of a surprise.
[07:56.230 --> 07:56.710]  That's cool.
[07:56.710 --> 07:58.090]  One of the other things I, you know,
[07:58.090 --> 07:59.310]  I've been hearing a lot about this stuff,
[07:59.310 --> 08:00.750]  but I was involved in a lot of production side.
[08:00.750 --> 08:02.570]  So I didn't know a lot of these details.
[08:02.910 --> 08:05.650]  When I first saw the air bearings in real life
[08:05.650 --> 08:07.850]  and understood, oh wait, that's how that works.
[08:07.850 --> 08:08.530]  That's what's happening.
[08:08.530 --> 08:09.270]  That blew me away.
[08:09.270 --> 08:10.310]  Can you talk a little bit more?
[08:10.310 --> 08:11.450]  You've got one there.
[08:11.750 --> 08:15.030]  Tell us about how that works and what's going on.
[08:16.130 --> 08:16.690]  Yeah.
[08:16.690 --> 08:20.910]  So the air bearing is like two precision pieces of aluminum.
[08:21.750 --> 08:24.250]  It really works almost like an air hockey set.
[08:24.250 --> 08:26.010]  Like we played with when we were kids.
[08:26.010 --> 08:26.770]  Like a table hockey.
[08:26.770 --> 08:26.950]  Yeah.
[08:27.410 --> 08:27.930]  Yeah.
[08:27.930 --> 08:28.190]  Yeah.
[08:28.270 --> 08:31.490]  So it has, you know, eight machines
[08:31.490 --> 08:34.550]  or eight precision holes in it.
[08:34.550 --> 08:37.030]  And you, you know, we hook an air compressor up to it.
[08:37.030 --> 08:39.330]  We set it to 10 PSI or so.
[08:39.390 --> 08:43.010]  And the top plate just floats on top of the air bearing
[08:43.010 --> 08:45.730]  and the satellites are sitting on top of that.
[08:45.910 --> 08:48.550]  A lot of work went into actually getting
[08:48.550 --> 08:50.090]  that to work consistently.
[08:50.410 --> 08:52.750]  And challenge two, we made the team's point
[08:53.310 --> 08:55.470]  at the sun for an entire revolution.
[08:55.470 --> 08:58.710]  And, you know, I've been playing with that for months.
[08:58.890 --> 09:01.610]  Air compressor in my office running consistently.
[09:01.770 --> 09:04.130]  So it was neat to see the teams do that too.
[09:04.130 --> 09:05.970]  And we didn't make it easy for them.
[09:06.290 --> 09:09.610]  So, so yeah, that's, that's essentially how it works.
[09:09.610 --> 09:09.790]  Excellent.
[09:09.790 --> 09:12.330]  Now, Jason Williams, how do you get air then
[09:12.330 --> 09:13.530]  on a thing that's moving?
[09:13.530 --> 09:14.970]  You have the carousel spinning around.
[09:14.970 --> 09:17.490]  What was, I mean, was that like a thing that you designed
[09:17.490 --> 09:19.150]  or built or bought or how did that work?
[09:20.390 --> 09:23.210]  So the center of the carousel is not moving
[09:23.210 --> 09:25.290]  and that's where we have the air hose come in.
[09:25.630 --> 09:27.630]  And then we have it hooked up to a manifold
[09:27.630 --> 09:30.030]  that distributes air to each of the teams
[09:30.030 --> 09:31.570]  on the carousel itself.
[09:31.570 --> 09:33.610]  And an interesting thing about the carousel is actually
[09:33.610 --> 09:37.390]  it's driven off of a slew drive, which is somebody
[09:37.390 --> 09:39.530]  if you're familiar with a solar tracking arrays
[09:39.530 --> 09:42.130]  the slew drives, what causes this solar tracking arrays
[09:42.130 --> 09:42.970]  to track the sun.
[09:42.970 --> 09:45.850]  So we actually got that from China, had it shipped here.
[09:45.850 --> 09:47.190]  It was kind of difficult to get
[09:47.710 --> 09:49.910]  but that's what the centerpiece you see in the room there
[09:49.910 --> 09:53.890]  that's rotating the carousel about 15 minutes rotation per
[09:53.890 --> 09:56.430]  sorry, 15 minutes per a single rotation.
[09:57.590 --> 09:57.950]  Awesome.
[09:57.950 --> 09:59.890]  Well, thank you both for taking some time out
[09:59.890 --> 10:00.570]  to discuss this.
[10:00.570 --> 10:01.150]  This was awesome.
[10:01.150 --> 10:03.090]  I'm sure the audience appreciates it.
[10:03.090 --> 10:05.590]  We'll let you go back to the game and we'll see our audience
[10:05.590 --> 10:08.690]  in another hour or so or next update.
[10:08.710 --> 10:09.930]  Take care.
[10:10.470 --> 10:11.330]  Thank you.
[10:11.330 --> 10:12.130]  Thanks.