A little over a month ago we announced our work developing an Augmented Reality Platform for the U.S. Government. While we can’t talk about contract negotiations as they are lengthy, sensitive, and tricky, we CAN share that our Bloom server is currently in an Alpha state and is already doing some amazing things. We are modeling, modeling, modeling from video and still photography and the output is amazing. The video above is a model of the front of my home. The video above represents only a single (but very important) step in the Bloom “Gathering Process.”
We are also in the process of optimizing the code to further improve efficiencies. As our effort continues we will be entering the second phase of our program that will include the development of the user interface for the iARM platform. That user interface will be a multichannel input / output leveraging gesture (hands/hand symbols), speech, and context. Our first evaluation will be the gesture based system. There are a number of open-source and closed platforms (this, and all of these…) that will likely play a part in this system but we are always looking for experts.
Please take the time to share if you have ideas you think would benefit this vision.
About a month ago I posted that we were making a MakerBot CNC Printer. Well, we’ve done it. After many evenings with a few of the more passionate robot-lovers here at Tanagram we completed the build and have begun running test prints. Building from a kit is a challenge if the instructions are vague and for the most part they weren’t but we always found 3 ways to interpret everything, as a group we worked well to decipher it. During our journey I roasted our first print head because I used an unconventional PID setting. That set us back a few weeks. We are now on our second print head and are working through calibrating the unit. It’s great to be printing things other than calibration blocks. You’ll see some photos below of some of our results.
Kim working on the chassis assembly.
Andrily working the small parts.
Plastic injection motor assembly made from laser cut and stacked plexi.
The handmade print head. A special wire is wrapped around a brass tube. When electricity passes it it heats up, it heats up FAST.
Another angle of the print head.
The body is made up of lasercut wood elements bolted together. You can see one of the Z-axis tractors bolts in the back corner.
Three motor controllers and a motherboard.
Kim inserting the X/Y axis and platform.
Build platform base.
I had to assemble and dissassemble print head frame 3 times because of a poorly glued idler wheel.
Testing the print head before mounting it. IT WORKS!!!
It spits out fishing line!!
Assembled and in its new home.
The first test print.
An awesome bear with an error (we call it poop) on his face.
Whistles that work! (the rear one was before we refined calibration)
Soon we’ll post lessons learned on this project. For now we are going to rest in our bliss as our little CupCake’s motors whirrrr away. We are dreaming of the things we will design to print.
I’m thinking I need a little fluff today, how about you?
Recently, I’ve seen quite a slew of examples of business cards that address questions I’ve always had about the value of the card as an object versus a ceremony. The results are gorgeous, but in a “different” way. Concepts and functionality make sure these cards *definitely* won’t end up in the trash can. What do you think?
Ramiro Pareja Veredas of Spain wanted a business card that would make an impression. His creation can be converted into a USB drive by punching out two of the corners. It can store data, works as an oscilloscope, datalogger and PIC trainer, and packs the CPU power of a Spectrum ZX. Click the image for more.
A laser-cut, punch-out "spirograph" art-machine. See more here.
Not the most exciting of the group, but fun at parties! Ok…no…
Designed and fabricated in a single night, their card contains all the parts needed to build a brushless DC motor. The only downside to the design is that you need a separate controller to actually make it run, however I'm sure that could be fit into a second card.
Combination business card and drawing template for organic chemistry
This makes sense for a mechanical engineer, but it looks like fun for kids and adults alike: make a business card (or scrap cardboard) into a small paper stock rubber band catapult!
Another great idea for a business card, Adrian's laser-cut card transforms into a handy little caliper.
On June 1st Jason Fried published “Never Read Another Resume” via Inc.com. It was thematically correct in the sense that Jason tends to go on irreverently about strange abstractions of reality. For me it was just another moment in his timeline. For Scott Olsen, it was the last straw. Two days ago Scott wrote a scathing rebuttal to Jason and all who follow him. Scott is a great writer and I found his views both hilarious and on-point until… he came after Chicago.
FUCK! – I AM SO TIRED OF PEOPLE SNIPING AT CHICAGO!
First, let me take care of Jason with a quick highly editorialized review of his journey. Ten(ish) years ago Jason and his colleagues, members of the elite design group 3rst (thirst), spun off a web-design company that they creatively named 37signals after the total number of signals SETI had captured that *could* represent intelligent life. They were successful as we all were because it was the very top of the internet bubble. It was raining gold those days.
Then the bubble burst.
Instead of joining a pottery club, or going back to school, or weeping at the edge of a valley that became a trailer park, Jason and his team decided to build a tool to help other design firms build closer relationships with their clients. They were successful, it’s called Basecamp. We use it, all of you should. Jason grew his self-proclaimed tiny empire out of the rubble and has since released a number of other successful products (some more than others). Jason changed his game and ours – for that and that alone, he deserves his spot in the sun. So here it is… Thank you Jason, you have made my business better and my clients happier. You are crazy, but I love you.
So why dig up Jason’s history in his defense? Because I believe you, Mr. Olson, have a Pernicious Lack of Perspective. If you know anything about Jason you should understand that he’s just a guy pointing at the moon, not giving point-by-point driving directions. He is organic, responsive, and chaotic. He is also brilliant. I fault you for taking his rants literally and also not maintaining your “west coast calm.” Jason is notorious for his ‘question everything’ approach. We share that in common but to a large extent he has found greater success in it.
Quickly, I have my share of experience with San Francisco and Silicon Valley. I love them both, they are entirely different. I’ve spent quite a bit of time in Silicon Valley, in offices with “recycled” furniture for office equipment. I’ve sat across from Jeff Hawkins (his space was nicer) sharing ideas on the future of intelligent computing. Sure, I’ve occasionally referred to SV as “The Land of Unicorns and Rainbows,” but it has always been with love. Who doesn’t love unicorns? It makes perfect sense that San Franciscans are very eager to annex SV, it’s a magical place.
So why am I based in Chicago? Very simple… Family. I was raised in Battle Creek Michigan and my wife grew up in Milwaukee. We are northerners. I don’t find solace in subtle seasonal shifts. I like change. I like leaves that drop and cold that bites, and snow that… well whatever snow does.. you get it.
I also love the people. Nobody hunkers down and get’s it done like Chicagoans. Sure we have a temper (you’re getting that, right?) and sure some of us have a funny accent, but we love what we do and we *are* “The City of Big Shoulders.”
Here’s a short list of comparisons of the two cities:
Remember that excited feeling you had at the beginning of a project where the sky was the limit and you were chomping at the bit to get started conceiving your next, greatest masterpiece? It’s now three-months later and you’re biting your nails raw in anticipation of the 3pm presentation of your concepts to your customer. It’s been a great exploration, you’ve done your best work, researched the heck out of the topic, and yet you are still praying they will love your ideas. What’s wrong with this picture? You may not know it but this situation is exactly the situation you don’t want to be in. The problem I’m illustrating is the conceptual disconnect that is causing you to lose sleep hoping they won’t dissect your concepts or worse combine them (not to mention that you are showing more than one concept). It’s the fact that the people sitting in the boardroom don’t know what they are about to see. Put bluntly, if your client is going to be surprised by your concept presentation deep into the project, you are risking failure. Sadly traditional methodologies encourage this pattern and the potential disconnects that often arise. To us, it’s just gambling.
How do we do it differently? We put our clients to work. A good friend of mine often refers to “Consultant’s Hubris” or the feeling that as a consultant we know more than the client. While we may be an expert at what we do, we hardly know the issues and opportunities the client faces on a daily (yearly) basis. We are not experts in their business, period. Face this and become more powerful.
Our contracts require client participation, sometimes significant hours weekly. When we kick-off we ask our client to designate a “steward” who will participate in the project heavily. The steward must have deep experience within the client organization and products we are working with. They are a Subject Matter Expert (SME).
We start with a working session to clearly define goals (or at least the first draft of the goals) and plan out how we are going to attack them. Following this are short 2-3 week work cycles (sprints) that involve twice-weekly working sessions where the client-steward and team are locked in a room discussing, researching, sketching and prototyping. Because our people are very experienced, they guide the steward’s ideas to merge with relevant trends and technology and result in beautiful solutions. At the end of each sprint we evaluate our collective success, present status to the larger client organization (gather additional feedback) and plan the next sprint. At the end of the project we have developed one concept with the steward that is not only extremely relevant but also co-owned. That’s right, the steward owns the solution because they co-developed it and they actually help ’sell’ our collective solution to their organization as we develop it. Of course this means they are circulating pencil sketches and prototypes well before the final presentation, but they are also translating the thinking into meaningful dialog with their colleagues. It’s very powerful to watch. The best part is we together are all heroes at the end and if you can make your client a hero, you’ve done your job correctly.
The first thing people ask me when I talk about this approach is how we handle the “bad” ideas stewards bring with them. The answer is communication. We’ve had stewards bring us wireframes of the application with full expectation that we would build exactly what they had specified. We embrace their concepts and talk directly about them with the steward. We explore the concepts that inspired them and the needs they are trying to solve. These concepts are immensely valuable windows into the mind our partner/client. During this conversation we share experience, industry trends, competitor approaches, and user-centric best practices. We do the drawing – together we are each the best at what we are. I’m proud to say that in every project we used this approach, the results have been stellar. The solution is something both we and the client are immensely proud of.
Because this method is designed to break down the walls between the client and our team it works very well with companies who have stakeholders with competing needs. Our process works very well in this “federated” situation because we incorporate stewards from each department in our working sessions. The debates happen in real-time and if they can’t be solved on the spot, they are resolved within the week with follow-up discussions.
There isn’t enough room on our blog to talk through every detail of this approach but if you’d like to learn more about this process we’re happy to help. Drop us a line or post-back here and we’ll be glad to share. If you like we can present our approach, train your team, and even collaborate on your next big thing!
ARE2010 did a lot of things for Tanagram. First and foremost it taught us that we aren’t the only crazy people out there. There were many hundred equally passionate (and equally insane) individuals working diligently to realize a vision that has been foretold for many many years. Attending were 400 enthusiasts, 90 speakers, and 40 sponsors.
The conference also re-affirmed that Augmented Reality (AR) is a new frontier. The discussions were widely diverse from immediate implementations of marketing gimmicks to future visions of integrated experiences and super-powered humans. Keynote speakers, Bruce Sterling, Will Wright, and Blaise Agüera y Arcas, and Jesse Schell were all brilliant. They taught us to be humble, focused, human-centered, and reminded us that this isn’t the first new frontier and to remember (and reuse) the past. Jesse Schell presented a humorous and sobering view of privacy and what AR could mean for it (his point was it was already too late…).
I personally had an opportunity to meet and discuss ideas with some of my all-time hero’s including Bruce Sterling and Blaise Agüera y Arcas.
On the technology front I can share that there is a definite future for head mounted display technology. We saw proposals for 150deg. FOV systems. It’s going to be a very exciting decade!
As soon as the ARE folks post the videos I encourage you to browse the sessions. I didn’t attend one that was not fascinating. The ARE event will be held next year in Santa Clara and we will definitely be there.
P.S. Through this experience I was introduced to Bruce Sterling’s Wired Blog – Beyond the Beyond. Bruce admits it is eclectic as hell and is really anything that interests him. He saw that as a weakness, but I see it as a strength. I really look forward to the “random” posts. They are always head-stretching. I also encourage you to review the twitter feed for the conference. Great stuff in there!
Simply put our “robot” is the brainchild of some of the brilliant minds contributing to Make Magazine that have started their own company (MakerBot Industries) producing open-source hardware robots that print three-dimensional objects (of just about anything) using ABS plastic (and a few other variants) or frosting (see ‘the frostruder’). It is a derivative of the RepRap self-replicating robot movement (also open-source hardware) and is gaining quite a bit of popularity among technology enthusiasts who want to experiment designing real-world objects.
Below you will find a photo documentation of our first few hours of progress. We’ll post more as the work continues.
The Unboxing... Lots of parts.
Hey Cool! A tube of skate bearings!
Kim found the belts!
The RepRap Motherboard (under a rainbow...sigh!)
The extruder controller board
The axis (X/Y/Z) motoro controllers
Kim checks, sorts, and detaches the lasercut wood components
This doesn't look anything like the picture of the parts provided in the kit!
The X-Stage subkit.
The Y-Stage subkit.
The Z-Stage subkit.
The body assembly subkit.
The X-Y Stage subkit.
The proud categorizers (Kim, Joe)
I think after we finish this we will need to build a rabid security bot to guard our cute little “cupcake.”
This is an exciting illustration of the power of “collective memory” a concept we use to describe the knowledge management-like capabilities presented by spatialized AR data. Depending on the amount of stored detail one could rewind a specific view through multitudes of perspectives of a single area to see information from yesterday or a few centuries ago. This concept is a key component in the architecture of our iARM ecology.
Because our system is vision based, we have an extensive amount historical footage for areas of interest and the daily use of our system continuously adds to our model making for an amazingly powerful captured dimensional recording of our world.
If we caught your attention with our last post but you’re having a difficult time understanding what’s so exciting about a little research and some images narrating a potential future rest assured we have a lot to talk about.
The first thing we need to address is one of the “Hows” of the visualization technology. Even though we have several patents pending, we are still too sensitive to expose the ‘whole tomato’ to the masses, but suffice it to say we have cracked the Augmented Reality visual registration nut. The following is our first delving into the details that are under development.
What is this nut? Let me explain…
If you’ve ever used the current lineup of “AR Apps” available on Android or iPhone platform you’ve seen a neat portal into the world of Augmented Reality. You typically see a camera view of the world around you presented on a hardware screen with digital information presented (most often clumsily) on top of the view. As you move the phone around you see that information move accordingly as if it’s locked to some physical location. Sometimes you see that information move or ‘drift’ for unknown reasons and that is really the problem with current tech AR.
Registration is the accuracy by which the computing platform can “lock” the digital world onto the physical world and up to today, that technology has been susceptible to all sorts of errors and interference. One of the reasons registration is still immature is because developers are relying on hardware improvements to solve their problems. We believe that while hardware, especially heavily funded hardware (thank you Apple), may eventually get accurate enough to solve the registration problem we don’t have the time or patience to wait. We instead spent our time researching and testing an elegant and simple solution that would work with today’s hardware capabilities.
If you look at the narrative slides below you can see our User Experience Model relies heavily on pixel-accurate digital registration. We can “paint” anything including highlighting objects, manually marking the environment and even embedded three-dimensional models. I can assure you these aren’t the crazy vision of a designer with no insight as to how the technology works but are based on our registration system (U.S. Patent Pending).
Our system is primarily vision-based, but leverages gross measurements (with error detection) of GPS, Cell/WIFI triangulation, and digital compass hardware. That means that it uses cameras, models, and some other fancy wizardry to rapidly identify the observed view and paint objects within that domain accordingly.
That’s all for now but we’ll be presenting some more details on the system at ARE2010 and are excited to continue this discussion.
About a year ago, Tanagram was granted its first direct (meaning we weren’t a sub-contractor) DARPA research funded program. Our goal was to improve situational awareness for soldiers deployed in contested (dangerous) cultures. During this program we proposed creating a complete server / client based architecture that responded to the program needs by providing digital augmentation of the surrounding environment. Tons of research, some prototype concepts, and a few patents later we are now cleared to share it with you.
We’ll talk more about the technology stack and client platform(s) in future posts, but for now you may read our final report.
Recent Comments