VI Shots

So Chris Anderson the Editor-in-chief of Wired magazine did the final keynote presentation at NIWeek 2007 this past August. I was really excited about this since up to the run up to NIWeek I had managed to purchase and read his book, The Long Tail. I was also eager to get him to sign it, which I managed to do.

Chris did an amazing job in the keynote of describing his involvement with the LEGO Mindstorms NXT robotics system and how he created a UAV which was controlled by the NXT. This is marketing gold for LEGO and an inspiration to all, young and old as to what can be accomplished with a few low cost LEGO parts and a cell phone. He mentioned how his 9 year old son programmed the NXT using the NXT-G graphical programming language (which we all know is really LabVIEW in disguise).

I left that keynote feeling happy and all geeked out because, hey, when it really comes down to it, the Editor-in-chief of Wired magazine is using LabVIEW! How cool is that?

Well, not so cool it turns out. Chris just posted on his geekdad blog that Robot C is “The Best Programming Language for Lego Mindstorms” I mean, c’mon Chris, really? Can you seriously tell us that a 9 year old kid will jump for joy when you show her a screen full of this? Give me a break. I’m beginning to wonder if you’ve turned from a geek dad to simply a geek. We shouldn’t lose sight of the fact that this intended for kids. No, really it is. The biggest challenge we have as parents is to find ways to pull our children away from various mind numbing distractions onto something useful. Showing them pages of C code is not one of them.

In this video my 8 year old son built and programmed the NXT claw on his own. I mentioned to him that he was programming in LabVIEW and he was surprised. He was surprise because he thought LabVIEW was this complicated software that only dad knew how to use and was for automating complicated machinery he knew nothing of. This is the whole point. He was able to build the mechanics and write the automation program without writing one line of code.

The future is not C or text based programming languages. The future is graphical programming. Chris mentions several times that it’s time to use a “real programming language”. I agree. This is why we should all be programming the NXT using LabVIEW. See, LabVIEW is a real programming language and contains all the things that Chris misses in NXT-G. LabVIEW has “if…then…else”, “while”, even “for…next”. It also has sophisticated debugging tools. Can it rise up to the challenge? Well, if Virginia Tech uses it for DARwIn and the DARPA challenge then I’m sure it will work on the poor little NXT brain. Oh, and if the cost is too high, there’s always LabVIEW Student Edition.

Edit: LAVA has picked up the discussion here.

It’s great to see LabVIEW once again at the forefront of autonomous robotics. This time with an impressive finish by Virginia Tech team Victor Tango and their vehicle named Odin.

Virginia Tech, along with TORC Technologies, won the $500,000 third place prize last weekend at the Defense Advanced Research Projects Agency (DARPA) Urban Challenge. In a close race with teams from Carnegie Mellon and Stanford universities, the Virginia Tech team used National Instruments LabVIEW software and CompactRIO hardware in its vehicle. Virginia Tech’s team, Victor Tango, was one of only six robotic teams to finish the 55-mile DARPA Urban Challenge course.

“National Instruments congratulates team Victor Tango on its remarkable achievement,” said Ray Almgren, NI vice president of academic relations. “Team Victor Tango is a great example of how domain experts, rather than computer scientists, use NI LabVIEW graphical system design to quickly design, prototype and deploy sophisticated robotic designs. NI is proud to offer technologies for applications in this exciting and growing field of mobile robotics.”

As part of the competition, TORC Technologies created a set of LabVIEW tools for Joint Architecture for Unmanned Systems (JAUS), an autonomous ground vehicle standard for passing messages and status information between various vehicle subsystems. LabVIEW running on a separate Microsoft Windows Server performed image processing and path planning. The team integrated an NI touch panel with the vehicle dashboard to select appropriate modes of operation.

“This exceptional team of Virginia Tech graduate and undergraduate students has been a true joy to work with, as they share the same passion for robotics as TORC,” said Michael Fleming, president of TORC Technologies. “With LabVIEW, the team implemented parallel processing of high-end vision algorithms running on two quad-core servers that perform the primary perception in our vehicle. The ability of LabVIEW to automatically multithread our application, in addition to the optimizations we performed in the language itself, drastically reduced our development time.”

Quicktime

Karl Muecke was at the Robo Development conference 2007 showing of DARwIn. DARwIn stands for Dynamic Anthropomorphic Robot with Intelligence. Karl was visiting from RoMeLa, the robotics and mechanisms laboratory at Virginia Tech. RoMeLa does a lot of research into robotics locomotion strategies.

The development of DARwIn started back in 2005 with version 1 and the current incarnation is at version 2b. DARwIn runs LabVIEW Real Time on a PC104 board and RoMeLa has decided to standardize on this platform for all their robots.

LabVIEW controls DARwIn’s motion over RS-485 and can read joint positions on the same serial network from the servo motors’ built-in potentiometers. While the robot is walking or moving, a rate gyro with acceleration and orientation information communicates with LabVIEW over an RS-232 serial connection so that the program modifies the walking gait to effectively balance the robot in real time.

Karl told VI Shots that RoMeLa is planning on evolving DARwIn hardware so it can be sold at the consumer level at a lower cost. I think this would be very cool.

View video in Quicktime format

Hanson Robotics has done some incredible work in the area of robotic human facial expression emulation. Check out their website which has several videos of their handiwork. David Hanson is a true pioneer and visionary in this field. With the invention of Frubber and special AI interaction algorithms, he is helping build the future of robots that can interact with the human population in a more natural fashion. I can’t help but compare him to doctor Noonien Soong who designed Data from Star Trek the next generation. A bit of a stretch, I admit, but listening to him speak in an interview at the 2007 Robo Development conference about robots that can “truly love” makes me wonder.

Hanson robotics is now getting into the consumer market with a creation they call Zeno. Zeno will be available in 2009 for the price range of $200-$300. Zeno is a robotic companion that can interact with you on an entirely new level that has not been seen before. It’s considered a toy targeting kids but I can see this being purchased by gadget craving adults as well.

Hanson Robotics is collaborating with Massive Software on this one. Massive has built their business on artificial intelligence algorithms used in computer animation for simulating crowds. They started off by developing the computer animation for Peter Jackson’s Lord of the Rings battle sequences.

The vision and decision making components in Massive Software give Zeno the ability to navigate, make facial expressions, and move his body based on what he sees in his physical environment. The video coming in from Zeno’s eye camera is fed into the Massive part of his brain so that he can move appropriately and respond emotionally to what is going on around him.