Hizook

This website will be transforming into a "personal" website, rather than the blog it currently is. I'm sure the new site will still have a blog component, but it will most likely be a complete overhaul.

Originally, I was going to switch it over to Drupal 6,1, but it appears that many useful modules (such as CCK and Views) are not available at the moment, so I will probably be migrating over to Drupal 5.7. This will be a lengthy process since (1) I'm really busy and (2) I need to build my theme.

Hopefully, all of my old posts will be migrated & maintain their URLs, but I can't make any promises... At any rate, all previous comments will be lost.

Catch ya'all on the flip-side.

People who know me well are aware that I enjoy good science fiction.  One of the best scifi books in recent history is Rainbows End by Vernor Vinge, which has won a 2007 Hugo nomination.

One of the things that makes Vinge a good scifi writer is that he is well educated in science and technology (former professor in math and CS).   According to Dr. Thad Starner, a wearables expert from GA Tech who is acquainted with Vinge, I know that Vinge was an early subscriber to the "Wearables Mailing List."  All of this means that Vinge is well-versed in the current capabilities and exciting future of wearable devices and human-computer interfaces.  To make a long story short, I love this book.

Anyway... I was thrilled to get an email from a colleague (Dan) that Vinge released a free, full-text, online copy of Rainbows End!!!  You can find the link on Vinge's website:  Vrinimi.org.  The direct link to the book can be found here.  Enjoy (and if you like the book, be sure to support the author by buying a copy -- we want to send a clear message that "we want more").

I'm happy that my parents understand my weird quirks. I've always had bizarre requests when it came to Xmas -- usually related to low-level electronics in some fashion or another. This year, I've asked for a contribution to my "Quest to Build the Ultimate Home Electronics and Computer Lab" fund; I plan to apply additional fellowship money to their gift to acquire a new addition to my personal lab...

I should probably update the "Quest to Build the Ultimate Home Electronics and Computer Lab" post by mentioning that I returned EZ Digital function generator and purchased the Instek one instead -- it was a much better piece of equipment....

Anyway, I'm struggling between a few possible choices:

	This is the Weller WD1000M single-channel soldering and re-work station. Basically, it is a very lightweight, fast-temperature-response soldering iron with digital temperature control. It is used almost exclusively for surface mount parts. I used one of these very briefly while at Sandia National Labs, and they are slick. Retail price: $445 (from HMC)
	This is the Instek GPS-3303 tri-channel digital benchtop power supply. It has two 0-30V/0-3A channels and one 5V/0-3A channel. It has 4 LED displays for voltage & current. Also allows for parallel/series operation for increased current/voltage. More details here. Retails for about $420 (cables included).
	This is the Mastech HY3003D-3. It is pretty much the same as the Instek model in that it has two 0-30V/0-3A channels, one 5V/0-3A channel, and 4 LED displays for voltage & current. I don't believe it comes with cables, so those add about $30 to the retail cost of $320 for a total price of $350 (from Jameco)

I guess another option would be a hot-air re-work station. Those fall in about the same price-range...

Anyway, what to do? Currently I do lots of prototyping -- some surface mount, but nothing too challenging. Though there are some high-density projects on the horizon, but I don't forsee any work being done on them in the next 6 months. I already have a decent WESD51 iron with some fine tips that are more than adequate to get the current job done. I also already have a single-channel 0-50V/0-0.5A analog supply -- which works for most things, though the analog display (requires DMM verification) is lacking, the 1/2 amp current rating gets me every so often, and I really hate having only a single supply.

I guess I'm kind of leaning towards a 3-channel benchtop supply. I can get a fast and accurate voltage/current readout. I can also use the three supplies to prototype analog circuits (5V for the digital, +-12V analog supplies). So I suppose the real question is... "Is the Instek worth the extra $70?" That's tough. I really like Instek's signal generator, and I'm comparing it to a relative no-name (at least for me). I was hoping I could get someone to weigh in with their thoughts...?

Bonus points for anyone who knows what this device is... I'll give you a hint, it's not a UFO (though it bares a striking resemblance).

OK, I just can't hold off the suspense any longer.  It is called a "pyranometer."  From the Wikipedia page:

Pyranometer clearly showing the instrument main components: glass dome, metal body, black sensor, radiation screen, level and cable. Dimensions: diameter of the dome is 40 mm. Photo shows model SR11 Courtesy Hukseflux Thermal Sensors.

These devices measure the incident solar irradiation by measuring the temperature change of an absorptive material (isolated by the glass domes).  They're used to tell you how much sun-light is hitting per square meter at any time throughout the day.  Luckily, a large dataset has already been compiled (worldwide) for many different locations in the TMY2 dataset.  All of this can be used to predict how much a PV system will generate in a typical year at a particular location, accounting for cloud cover, fog, etc.  For Atlanta (my residence), that works out to about 4.5 kWh/m²/day if you take the daily average. 

Of course, it might just be easier still to just use a NREL chart.

Anyway, I had never heard of a "pyranometer" before...  Good to know.

I saw a presentation by Professor Siciliano from University of Naples a few weeks ago. During the presentation, he briefly mentioned (and had a video) of a very cool new humanoid named Justin.

Justin sports not one, but two of the DLR-III Light-Weight arms (now being manufactured by Kuka).

These arms are impressive! They're the first robotic arms (to my knowledge) to have a 1-to-1 power-to-weight ratio (meaning they can lift their own mass). These are very "sexy" arms, and even more impressive in person (Georgia Tech has one in orange that I spent some time examining -- way cool).

Anyway, I was unable to find copies of the videos Professor Siciliano showed during his presentation. The best I could find was a video on Wired's site from ICRA-2007. They apparently do not allow downloads, so the following was the best I could muster: Justin-ICRA07

Perhaps the ultimate in wearable computing and humanoid robots is an exoskeleton. If you recall, I was privaledged to watch a talk given by Stephen Jacobsen of Sarcos at the Wearables conference. I tried finding copies of the images and videos he showed, but at the time, they were closely guarded... Well, now Engadget has posted similar to the ones Mr. Jacobsen showed us a few months ago...

So here is a picture, followed by a video (local copy here).

It's been more than two weeks since my last post, and I'm feeling guilty about this website, so I'll amuse everyone and fill people in...

I've spent the last week placing parts orders for a new robot (well, a less-capable version of one we already use). Two large components of the system are the Erratic base by Videre Design and a SICK LMS 291-S05 laser rangefinder (both shown below).

I also acquired two Hokuyo URG-04LX laser rangefinders from Acroname. It is surprising how much of "industry" is littered with registration forms and "call for quote" systems. It is infinitely frustrating! I'm glad that places like Videre and Acroname make life easier by having product prices listed (sans registration) and allow easy online ordering.

Let's see... I've also been really busy with classes. We just had our last round of exams before finals, so that was trying. We also had a term paper due for "Solar Cells." The topic was of our own choosing, and I chose "Inflatable Membrane Solar Concentration Systems for Space-Based Applications." How's that for a title?!? I've actually uploaded the paper (since I'm rather proud of it) here. One of the cooler technologies I talk about was developed by Cool Earth Solar, which was founded by a former Sandian from Livermore, CA (my former employer). Their idea is to use low-cost, inflatable concentrators and high-efficiency PV cells to tackle the $/Watt challenge for terrestrial systems (also useful for space-based systems). Cool idea. Check out a few figures from my paper...

In a weird twist of irony, a friend-of-a-friend at Georgia Tech was one of seven interns at this small (less than 10 full-time employees) company last summer. Small world!

So let's see. I've also been reading a lot.  I set a goal for myself to watch less TV and read/write more.  I've accomplished that goal in the last two months by finishing five science fiction books, many of which are classics I've always procrastinated reading!  The list includes:

• Time Machine by HG Wells
• Rendezvous with Rama by Arthur C. Clark
• RAMA II by Arthur C. Clark
• Childhood's End by Arthur C. Clark
• Schismatrix Plus by Bruce Sterling

I've also been reading a lot of journal articles.  Some have been focused on literature surveys (some areas I won't mention right now -- don't want to give away anything).  However, one area is in artificial robot skin -- I think I may take "Multidisciplinary Robotics Research" (a project-based class) next semester to develop my own, new twist on robot skin.  I've also been reading journal articles on materials science, there's so much cool stuff, and not enough time to talk about them!  Which brings me to my next point...

There just isn't enough time to work on a top-notch blog by oneself (at least not while trying to take grad classes, do cutting-edge research, read copious books/literature, and try to author a book myself -- a story for another day).  I do, however, enjoy writing good articles (much like the Solar Cells term paper), and I certainly enjoy meandering online exploration -- oh, the countless hours of "wasted" time. 

With this in mind, I've been speaking with a dozen-plus friends/colleagues about writing bi-weekly, quality articles, doing webhosting and webdesign, etc. for a "combined" site.  Right now, it's really just a pie-in-the-sky idea (one that has been postponed due to cancelled travel plans), but I really would like to see it happen.  Given the expertise of those around me, the site will probably focus on robotics and related topics (the 'related topics' include almost any scientific topic, from biology, MEMS, materials, sensors, chemistry, user-interfaces, etc -- given the multidisciplinary aspect of robotics).  Let me know your thoughts and if you'd like to contribute -- just be warned, there is probably no fortune and very little fame to be found in the endeavor.  

And with that, we're back to robotics -- what I should be working on right now!  So with that, I'll sign-out.

A few quick news updates.  First, the "scary." 

Atlanta only has 80 days' worth of water remaining.  The Red Cross is already suggesting stocking 5 cases of bottled water per person to last for two weeks if/when the shortage becomes hyper-severe.

Though I was never overly involved with Scouts, I believe their "Always be prepared" motto applies in this case.  Fei & I will have to stock up on water.  I've also been meaning to buy a hand-held water purification pump for camping, so maybe I'll grab one of those too (it's timely). 

The drought is quite frightening for a number of reasons.  Obviously, it would be a huge crisis no matter the location, but this is Atlanta!  The people here are (by and large) crazy -- I can only imagine what it will be like during an emergency.  In Nebraska, or even San Francisco, I could see communities stepping up to work together to solve a problem.  Here in the ATL, people (on average) are very self-serving... it would be chaos.

Next, the "bad."

I had originally planned to purchase a Mac Book Pro in late-October or early-November.  I've decided this is not really an fiscally-intelligent decision given that I've owned my awesome Fujitsu tablet for less than one year.

I am thrilled about the release of OS X 10.5.  Everything I've read and seen about that OS screams awesomeness.  I think they are still working out a few bugs (like Time-Machine remote backups), but it is an awesome piece of work!

In the meantime, I may follow Charlie Kemp's lead and convert my laptop over to the new Ubuntu Gutsy Gibbon.  He has full tablet support functioning, along with free PDF annotation software (something I wish was available in Acrobat Reader or Foxit for free). 

Also on the "bad" front is the fact that the US Dollar value has fallen to the point where $1 USD is equal to $0.95 CAD (Canadian Dollars).  Craziness!!

I'm working on two new papers due early next year.  I'm really excited about both and will be working extremely hard to complete both by the deadlines.  It's nice to see everything come together in terms of empirical and theoretical.  On a related note, I highly recommend the Probabilistic Robotics book by Thrun, Burgard, adn Fox.  It should be required reading for anyone interested in modern robotics.

And now, the "best," as a little bonus.

There are some serious rumors flying around the internet about an Apple Tablet!

This is great news, given my "bad" news earlier.  If they release it in the next few months, then by the time you consider the time-padding for bug iron-out, I could buy an Mac Book Pro Tablet sometime next year! 

I really do like the tablet functionality, and I can't imagine mobile computing without it.  I know that sounds pitiful, but every tablet user I've ever spoken to says the exact same.  Anyone who refutes this has probably never used a tablet for a prolonged period of time...

OK, that's all.  I need to quit slacking off and get back to work!

I'd like to make a quick note about the Georgia Tech DARPA Urban Grand Challenge team, Sting Racing.  Unfortunately, they did not qualify for the finals, do the dismay of all of us who do not get to watch the competition live.

You can learn more about the failure of the robot on the team's blog, but essentially,

In our Saturday afternoon test in Area A Sting 1 crashed head-on into a concrete barrier. The front sensor mount was bent severely and pushed into the front of the vehicle. Fortunately the protection offered by the design and strength of the sensor mount saved the sensors.

After assessing the log data we determined the cause of the accident was a failure in the communication link between our GPS/IMU and the main control computer. Without “pose” information the robot could not know that it was moving.

If you're in to robot carnage, you can watch the video below.

OK, time to get back to work. 

UPDATE 11/29/2007:  I found a copy of a poster that went to RSS-2007 for this project here (local copy here).  

During the Spring 2007 semester, several friends (and labmates) took a course at GA Tech on mobile manipulation. This was no ordinary class... they used a Segway base with KUKA arm to fetch a cup of coffee!

There are a ton of reasons that this is interesting, from mobility, navigation, manipulation, etc. However, the most impressive thing is that each group used different software to complete the task. If memory serves, one used MS Robotics Studio, another used Player/Stage on Linux, and another used a functional language called OCaml on Mac (right?). Some probably question whether or not it works, well it does! And here is the video proof. From Henrik's blog,

The exam/demonstration for 4632B on mobile manipulation took place today. One group did 3 complete cycles of coffee delivery, the two other groups did partial cycles. The system is composed of a Segway RMP200 with a KUKA KR-5 sixx arm mounted on top. The strategies demonstrated included - particle based object localization, MC localization, visual servoing (image and 3D based).

I'm really pissed that I didn't take this class. I didn't even know it was being offered, and they ended up working on it about 30 feet from my desk.

OK, there are numerous competitions for aerial robotics.  Usually they are winged or rotary (helicopter).  Well, now NASA has launched the Lunar Lander Competition (Engadget discussion), where you can vie for a $2 Million prize.

A description of the task from the NASA website.

To win the prize, teams must demonstrate a rocket-propelled vehicle and payload that takes off vertically, climbs to a defined altitude, flies for a pre-determined amount of time, and then land vertically on a target that is a fixed distance from the launch pad. After landing, the vehicle must take off again within a predetermined time, fly for a certain amount of time and then land back on its original launch pad. There are two levels of difficulty, with awards for first and second place at each level.

OK, that sounds much like other aerial robot competitions, except... "rocket-propelled."  At this point, bells, whistles, and alarms were going off in my head.  From my model rocket days I recall it being illegal to add guidance systems to rockets, less they become missiles.  While there are no direct laws affecting model rocket guidance systems, the National Rocketry Association (and the FBI that watches them closely, and several states' court rulings) essentially say that they are illegal.  

So my question is... assuming I had the time and desire to participate, doesn't this give me legal, free-range to develop a "guided missile?"  It's a good thing I have little of either, as I'd hate to explain to the "man" that "NASA told me I could build a missile." 

I was frustrated yesterday when I couldn't directly download the videos of the dirigible and submersible manta ray robots from Festo.  It appeared that they were loading their own Flash player, which subsequently loaded the video -- all transparent to (and thus not "savable" or cached by) the web browser.  This seems like a problem that others may have (yeah right!), so I'll post my solution.

Not wanting to give up, I contacted the YouTuber who managed to extract and post one Air_Ray and one Aqua_Ray video.  His "I looked at the actions" response was sufficient for me to deduce a FLV video file recovery procedure.

The key thing is to determine the actions taken in the SWF file.  SWF is a compiled file format, meaning it is entirely not human readable.  Getting it in a format that can be read is a simple matter of decompiling the SWF file (using SoThink's SWF Decompiler, Demo Version).  Thus, the SWF file from:

http://www.festo.com/INetDomino/files/Aquaray_com.swf 

Shows an action script:

// Action script...

// [onClipEvent of sprite 9 in frame 1]
onClipEvent (construct)
{
    autoPlay = true;
    autoRewind = true;
    autoSize = true;
    bufferTime = 1.000000E-001;
    contentPath = "Trailer_Aqua_ray_2007_Bionik.flv";
    isLive = false;
    maintainAspectRatio = true;
    skin = "http://www.festo.com/INetDomino/files/Steuerung_Festo.swf";
    skinAutoHide = false;
    totalTime = 0;
    volume = 100;
}

And I can then download the FLV video from:

http://www.festo.com/INetDomino/files/Trailer_Aqua_ray_2007_Bionik.flv

Slick!  In this way, I am able to download the actual videos for my own storage.   Like I mentioned in the comments section of the Manta Ray post, I should also mention why I desire copies of the videos:

1.) To keep them for my own vices (mainly posterity -- stuff on the net seems to disappear over time, just look at Troody).

2.) To reduce transfer bandwidth from Festo's web-hosting.

I have no desire to infringe on their copyrighted videos.  I will see no profit by saving the videos, and will not use them without proper attribution...

UPDATE 10/23/2007: I managed to acquire the videos (see this post).  You can find the higher-resolution, permanent videos here, here, and here.

UPDATE 10/22/2007: I found two of the videos on YouTube, thus I have permanent copies of those two here and here. I'm still trying to figure out a way to obtain the other 1 (and some of the other ones from that site for future posts).

Festo makes some pretty awesome automation equipment. Besides "typical" engineering work, they apparently do some fun research applications. Today I'm going to show their robotic dirigible and submersible manta rays. Before discussing the details, take a look at their life-like movements!

The dirigible is my personal favorite (namely because it doesn't require scuba gear to observe), but both look very graceful. I can just imagine what it might be like on an alien world watching light-weight aerial "manta rays" swimming in the sky -- it would be amazing!

Air_ray, modelled on the manta ray, is a remote-controlled hybrid construction consisting of a helium-filled ballonet and a beating wing drive. Its light design makes it possible for it to “swim” in the sea of air, boosted by helium, in a similar way to the manta ray in water.

Propulsion is achieved by a beating wing drive. The servo drive-controlled wing, which can move up and down, utilises the Fin Ray Effect® and is based on alternate pulling and pushing flanks connected via frames. When pressure is exerted on one flank, the geometrical structure curves automatically against the direction of the influencing force. A servo drive pulls the two flanks alternately in the longitudinal direction, thus moving the wing up and down.

Aqua_ray is a remote-controlled fish driven by water hydraulics, the shape and movements of which have been based on the model of a manta ray.

The central drive and control unit of Aqua_ray takes the form of a Festo Fluidic Muscle. This is combined with the Fin Ray Effect®, a design based on the functional anatomy of a fish’s fin that makes it possible to imitate the fin drive of the natural role model almost perfectly.

As the Aqua_ray can be manoeuvred extremely well, and can be operated both as a hydrostatic glider and with an active wing beat, substantial energy savings can be achieved. Thanks to its shape and method of movement, the Aqua_ray can be used in wide ranging areas of oceanography, without disrupting the natural environment.

Robert Bussard passed away earlier this month. This is a huge tragedy in the light of his Polywell-Fusion research. His Google Talk a while back (which I spoke about here) was quite spectacular and inspiring. I hope that someone will pick up the slack and carry on his research.

I still recommend taking a look at his talk...

I love ImageMagick. You can use it to resize animated GIFs as follows (since a normal 'convert -resize' is insufficient).

convert input.gif -coalesce coalesce.gif

convert coalesce.gif -resize 300x150 output.gif

Slick, yes? Of course, you can resize to any resolution you please...