Lego Robotics – Meh.

July, 2012

While on holiday with my family in Orlando, FL we visited Legoland.  I made an impulse buy of the Lego Mindstorms NTX 2.0 kit for about $650 USD.  I later regretted it since I was unable to fit it in my luggage for the trip home.  I’ve been a huge fan of Lego since I was about 9 years old and I know their stuff is great quality.  I’m also enjoying robotics as a hobby, so I figured the Lego robotics platform was a must-have.  Also, First Lego League (FLL) competitions were something I was considering for my Robotics club with younger kids, but I wanted to get hands-on experience with it first before I committed to using it in an upcoming class.

NTX 2.0

Upon opening the box, I was greeted with the familiar bags of parts and no manual!  I later discovered that the manuals are all digitized and I needed to access them online from Lego’s (or a fans) website.  I familiarized myself with some new parts including the controller and some sensors.  I marveled at the engineering involved with the packaging of these devices to make them modular and fit into the highly standardized Lego footprint.  As expected, there were various options for building so I knew this investment would keep me busy for a while.  I was not familiar with the NXT 1.0 platform, so didn’t have any reference to compare the improvements made in this second generation.

First Build – ShooterBot

I decided my first robot should be the tracked platform with ball shooter.  I opened the required bags on the floor and awkwardly set up my laptop so I could follow the online instructions.

Lego Robot 2

I would consider myself an advanced Lego builder having owned and built dozens of kits including some of the advanced Lego Technic kits.  I selected this design thinking it would be a moderately complex build.  However, about half way through the build, I found myself lost.  Perhaps it is due to my colour-blindness, but I really struggled with some of the steps.  After building – undoing – rebuilding a few of the steps, I finally got the build complete.

The next step was the programming.  I installed the software developer kit on my laptop and followed the “Getting started” guide.  I was pleasantly surprised to learn my laptop had Bluetooth and was able to communicate to the NXT controller wirelessly (I’d never had a need to use bluetooth on my laptop previously).  I downloaded the Shooterbot.rbt file and loaded it into IDE.  I examined the “program” to understand what it was doing, then uploaded it to the robot.

I tested the robot by running the program and enjoyed launching the colored plastic balls across the room (and at my kids).  I think I lost 2 of the balls during the process since I was unprepared for the force and distance that the balls were launched!  I went back into the “program” and made some adjustments to make the robot do things a bit differently.  I certainly appreciated the power and ease-of-use of the IDE – I can see how younger kids would be able to use this effectively on their own after some instruction.

Second Build – Humanoid

Ok, time to move on to something bigger and better.  I selected the most advanced option available – the humanoid robot.  It used almost all the parts provided in the kit, which was great.  It wasn’t long before I was lost again with the instructions.  This time was much worse – it took me quite a while to complete this build.  A lot of the joy of building was lost on me since most of the time, I had no idea what part of the robot I was working on assembling (hmm, could this be a leg…?).  It was only toward the end of the build when major parts came together that I was able to appreciate how it all fit together and the design considerations for the build.

Lego Robot

After the build was complete, I downloaded the “program” from the internet and uploaded it to the robot.  Frankly, the frustration of this build really knocked the wind out of me on this and I was losing interest fast.  I’ve never had that happen before in a Lego build!  I spent minimal time reviewing the complex-looking program to understand what it was doing.  The final result is impressive – the humanoid robot stands a full 13 inches tall and weighs more than most of the Lego creations I’ve ever built.  The movements and use of the sensors is very impressive – he walks smoothly and the design team did an excellent job with the final product.

Conclusion

Overall, I’ve decided not to move forward with the Lego Mindstorm NXT 2.0 platform for my Robotics Club (targeting 5th – 8th grades) for the following reasons:

  • The price is way too high
    • I try to target a price of $100 – $125 for the club robots so that each student can have their own and can take it home with them after the class
    • With this price, I would only be able to buy a few kits, and the students would need to work in teams.  Yes, yes, I know that working in teams is an important part of learning, but I feel strongly that less learning happens with that approach (especially when introducing topics for the very first time).  Also, the students would not be able to take the kits home for continued experimentation.
    • If I charged each student for their own kit, I would very likely not achieve the minimum class size to run the program
    • Most of my robots leverage open source hardware and software which make them much more affordable
  • The builds are very complex
    • If I got lost, I was concerned that my students would get very lost.
    • I would need a lot more proctors in the class to help them complete the build within the time allotted.
  • Lack of exposure to electronics
    • I teach in my classes that there are 3 main disciplines to robot building:  Software, Mechanical, and Electrical.
    • Mechanical:  Lego has always excelled at the Mechanical stuff – no question.
    • Software:  This was my first exposure to Lego’s Software and I was impressed.  There’s minimal typing required – just a lot of mouse activities which is good for younger kids.  It lets the kids focus more on the logic and less on the syntax which I think is a positive thing.  There is a considerable learning curve with this IDE, but once that is passed, I feel young kids could successfully complete a build with minimal supervision.
    • Electrical:  There is zero exposure to electronics in this kit.  Lego has done a great job of making the kits modular and easy to assemble, but in the process, they have completely eliminated the third discipline of robot building.
    • In my classes, I insist that each kid get exposed to a taste of all 3 disciplines.  I admit that I’ve struggled with making the Software exposure meaningful, so still have some work to do there.  However, the feedback from some of my students is that their favorite part is cutting and stripping the wires.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s