Friday, June 3, 2011

LEGO Engineering Symposium 2011 - Part 1

Reflections on LEGO Engineering Symposium 2011 (Tufts University’s CEEO, 24-26 May 2011) - Part 1 of 3.

Okay, so I confess, I’m a CEEO groupie. I’ve been a CEEO groupie for nearly a decade. Ever since I fired off a series of emails to the only address I could find when I wanted help with how to do something or other that was at the edges of what was possible with Robolab. I was most impressed that the person, Chris, who answered my emails was not only prepared to give me so much of his time, but was also pretty knowledgeable about the RCX and Robolab. Eventually, wondering if fielding Robolab queries was even part of this guy’s job, I thanked him for his time and asked what his role was in relation to Robolab… he replied that he wrote it! Only in his spare time, mind you. Chris’ main focus was particle-laden air turbulence, but he had wanted to use the Lego Mindstorms RCX to help teach engineering principles to undergrads. At that time, only Windows-based software was available for programming the RCX, and being such a Macintosh tragic, Chris decided to write some Mac-friendly software himself!

I think this story highlights exactly what it is that I love about the CEEO. Not only do they come up with creative and fun ways of helping people of all ages learn how to think like engineers, but they are always open to feedback and incredibly supportive of other people’s idea. They are genuinely committed to learning, and understand that their products are the means, not the ends.

I attended my first LEGO Engineering Symposium last year, and as much as I loved it (which was a lot), this year’s was even better! Everything I loved about the format of last year’s event worked just as well this year, but the big improvement was what it didn’t have… Last year the tipping point / climate change / renewable energy theme, undoubtedly an important issue, felt unnecessarily forced at times, particularly in the development labs. This weakness was recognised by the organisers, and full credit to Merredith for putting together a program that allowed the CEEO staffers (collectively known as the “red shirts”) to work to their strengths. A balanced program no-less; one that examined not only the “big ideas” of research, but also the pragmatic questions of “what am I going to do with my class tomorrow?”.

Given that most of the material presented throughout the Symposium is available from the LEGO Engineering website, I’m not going to try to summarise it all here. Instead I’m using this an opportunity to reflect on anything that took my fancy, anything I was particularly surprised by, or anything that I particularly want to remember for the future.

Presentations (Day One) – 24 May 2011

Merredith Portsmore - Introduction

Merredith explained that being a Symposium (as opposed to a conference) means that the event is more about sharing and collaboration than training. Hence we’ve got “development labs” rather than “workshops”. It’s also an opportunity to provide feedback to developers, in particular LEGO and National Instruments.

David Hammer – Listening to students
David was in form. He started with a repeat of the “juice box” and “dorm room” clips from last year, but this year took it a step further and in five minutes covered what took me weeks to wrap my head around when I took his class last Fall. Here are the main beats of his presentation…
  • A view of science…
    - Science as a pursuit: Of coherent, mechanistic accounts of natural phenomena.
    - Science as a body knowledge: The accounts that have result of that pursuit!
  • Two well established findings…
    - Children have extensive intellectual resources for learning science.
    - High school and college (university) students typically treat science as information to memorise.
    Why the second, given the first??
  • A likely conjecture
    - We assess ideas, and teach students to assess ideas for alignment with the canon — the results of scientists’ inquiries — rather than by the ideas’ merits within the students’ inquiries.
  • We need to change...
    - We need to assess students’ ideas — and teach them to assess their ideas — in ways that build toward how scientists assess ideas.
    - And that means a kind of caring about students as nascent scientists.
I loved his quick example that if, for example, I told you that, “I walked my dog to the park” and you replied, “Oh, don’t you mean ‘aardvark’?”, it would be a bit odd of me to agree, “Oh, yeah, I walked my aardvark to the park” as if it didn’t affect the meaning or that it even mattered.

David compared learning in science to what engineers do, putting ideas together and taking them apart.

Rob Linsalata – URAPI: Universal Robot API
Rob raised a very cool idea that I was certainly aware last Fall, but at that stage it was very early days. It looks like quite a bit of progress has been made since I left the Center.

At present there are a number of different hardware platforms that are targeted at the education market but have different strengths. For example, the modularity of Mindstorms that means that everyone’s robot can be different. There’s the simplicity of the Finch that means hardware does what’s it meant to do so the student can concentrate on programming. The iRobot Create that makes the base of a robust, commercially available robot chassis available as a building platform. (Some other platforms I’d love to see added to the mix would be NAO , LynxMotion, or even Willow Garage’s PR2 - when the price drops a few orders of magnitude...)

Imagine being able to run the same program on multiple platforms. Wouldn’t it be wonderful if they all played nice together? Having learned basic programming techniques on the Finch, the student might use the same code to program an NXT that they had built themselves. Or better still, I could imagine constructing a LEGO attachment, say an arm and gripper, on top of the Create base and then using one program to manage sensor information from both sections to control both LEGO and Create actuators.

As an aside, I recognise that having them play nice together might appear to work against the developers’ commercial interests. I.e. what’s in it for LEGO? Well, for a start, I’m a teacher so I think that learning trumps commercial interest, but second, I think there’s plenty of evidence to show that in communities where the players play nice together, everybody is better off. Think of cars manufacturers… what if learning to drive a Subaru was completely different to learning how to drive a Toyota? There’s no doubt that both companies benefit from operating successfully side-by-side in the same environment and enabling users to switch between them.

In practise, though, differences between platforms mean that different platforms will necessarily mean slightly different programs, but at least being able to use one software environment for multiple hardware platforms would be an excellent step forward. And I’m happy to see it happening within LabVIEW.

John Heffernan – Rewards and challenges of designing a PK-6 engineering curriculum
John presented his approach to a PK-6 engineering curriculum that makes use of BeeBots, WeDo, and NXT.

There was a lot of interest in John presentation and it was amazing how often it would be used as a reference point in conversations about a whole bunch of topics.

One challenge that I thought John described very nicely was the difficulty in finding suitable challenges for the youngest grades that would hit the right balance between being open-ended and achievable.

I liked the idea of having students “write down three ideas” (words and/or pictures) to help the students pause for a moment and think about their options when they’re get started with a design challenge.

Challenge ideas:
  • BeeBots - Find the honey
  • WeDo - amusement park rides, burglar alarms
  • NXT - robot car challenge
Make a robot that goes as fast as possible, but avoids obstacles – this seems like such an obvious challenge, but it never occurred to me as a way of dealing with the request I get every year from students who just want to make dragsters.

A few other nuggets:
  • We should be “engineering the curriculum”.
  • Do adults follow the model?
  • The challenges are the rewards.
John posted some of his "take away" ideas from the symposium here.

Pernille France (LEGO) – Pushing the envelope of engagement with LEGO

I missed this one because I was setting up for our development lab… (-:

Bran Gravel – Diverse Trajectories: Students' Multiple Representations and Varying Ways of Developing Understandings
In a slick presentation, Brian argued that the “language” of STEM is representation and the representations are central to the STEM disciplines. Representations range from the conventional to the idiosyncratic

Multiple representations...
  • Students’ resources - their ideas - are captured in different representations
  • Engage students with different aspects of problems
  • Offer opportunities for re-description
  • Provide opportunities to assess the quality of ideas
Brian provided some neat examples of what we might learn from students representing ideas in different ways, and how these ideas grows as they refine their representations. Students ideas are embodied in representations and representations can amplify efforts to make sense of the world.

John Cole (Dexter Industries) – New developments in sensors for the Mindstorms NXT
Dexter Industries have produced some very interesting new sensors for the NXT over the last couple of years. I was somewhat surprised to learn that they are such a new company. For some reason I had assumed that they were an established company that had only recently turned their attention to the NXT.

In any event, they have already got some pretty cool sensors on the market, and have some interesting new ones on the way, including:
  • Thermal infrared
  • WiFi
  • IMU (gyro and accelerometer combined)
Some possible IMU activities…
  • Centrifuge test
  • Conservation of momentum
  • Building a better bumper – crash test dummies
I’m particularly interested in the possibilities of using their dGPS and NXT-Bee sensors in tandem to support an NXT-based robot swarm.

Later in the day, I introduced myself to John and was pleased to have the opportunity to express my concerns about the name “NXTBee”. When I first saw the name I was troubled about its similarity to the name “N-X-Bee” that my son and I called one of our robots a few years ago. John was pretty understanding about it, and I’m confident that we’ll be able to keep matters out of court. (-:

Seriously though, John is a really nice and approachable guy, and made it very clear that he is interested in listening to suggestions for products and future directions. All in all, I think that what he’s done provides a great role model for anyone interested in carving out a niche in the market.

Karl Wendt – High Tech High Media Arts Engineering
Karl showed us some of the work he’s been doing with his students at High Tech High Media Arts. I’m not sure what was more impressive... how quickly he was able to put together the infrastructure and materials to create a program that enables students to build robots from scratch, or the quality of the students’ documentation of the process. He has made excellent use of Google sites to allow the students to document and display their work. Karl presented well, but ultimately I think the work speaks for itself… http://www.masei.org/1st-semester-senior-projects. Most impressive!

A couple of neat ideas…
  • Embracing failure – in bringing an idea from conception to the final projects, failure is inevitable and may lead to discoveries that are useful at another time.
  • Legacy – Karl’s students are required to address the question of what legacy will their projects leave behind. At the very least, past projects are kept as part of a 3D reference library.
When I tell people that I spent six months in Boston learning about robotics and engineering education, I’m sometimes asked, “where are you going next?” I think I’ve just found it! At the very least, I’ll be very keen to arrange a visit the next time I’m passing by.

Marcelo Molina – Robotic Education as a Tool to Help Change Developing Countries: Brazil’s Perspective
Marcelo’s presentation reminded us that students have much gain from an engineering approach to learning wherever you happen to be in the world. Talking to him to afterwards it was very clear that he faces some distinctly local challenges in doing what many of us (at least in the US and Australia) would take for granted. In spite of the challenges, he’s had some very impressive growth with his program.

Ideas for challenges

2 comments:

  1. Rob, thanks for mediating between Dexter Industries and your son. We are glad we could settle this out of court!

    ReplyDelete