Jul 14, 2008

NXT Campers build Sumo Robots

Here are some highlights from one of my NXT camps that concluded last week.

I had 5th - 8th graders with no prior experience with the NXT. We started off building the Tribot from the NXT Education kit and quickly moved to basic dead reckoning navigation.

The students were then required to derive axle rotations from a measured driving distance.
( r = d / wheel circumference )
In addition, they were required to derive axle rotations from an expected vehicle turn angle.

We then moved on to other forms of navigation using the light , touch and ultrasonic sensors. By the fourth day, the student were able to develop solutions for the Sumo challenge.

On day five, the students documented their work with team web pages including programming and LDD screen shots and downloads. I've listed the team web pages below along with some video taken from the Sumo challenge and practice runs.


Team Web Pages
[ Desctructo-Bot ] [Dexter] [Kill Bot] [Owners Of All ] [Terminators]

Sumo Challenge

Sumo Testing


Robolab 2.9 said...

Hahahahs, great job David! I did the same project for a course I taught back in June, but with the RCXs. 12 3-5th graders, trying to Sumo, and practicing Gracious Professionalism, it was priceless. We did not do so much of the dead-reckoning maths, but we did talk about gear ratios, inertia (in the Acrobot/flipbot), traction, etc.

Well done again!


David Levy said...

Studying gear ratios and traction would be great additions to the Sumo challenge or maybe even offered as an advanced curriculum.

Thanks Richard!

Robolab 2.9 said...


As it was for slightly younger kids with little time (2 hours, 4 days), it was a simple introduction, but I think they really learned something.

For gear ratios, all we told them was that having the same sized gears would make the robot drive at the "normal" speed. Different sized gears would make the speed and pushing power (we mentioned torque, but just called it pushing power) different. They were to experiment to see which one they liked, and had to base their Sumo strategies on it... fast robots usually were trying to hide from the opponent, strong ones trying to push. Since they only had two motors, one team actually tried to use the gears to build a ramp that moved up and down. It did not work well, but it looked great for the parent showcase. ;-)

Traction... same deal: we had them try different wheels on the tankbot, and observe the speed and torque.

So, a lot of my course was based on experimenting, rather than straightforward teaching. What about you David? How do you get your information to your children? Anyone else?


David Levy said...

NXT-G from LEGO Edu comes with a pretty good set of tutorials that cover individual strategies like "stop on black line" , "follow line",
"point turn" , "detect distance" ...

The kids were encouraged to borrow from those exercises when coding challenge solutions. I also spent a good deal of time making sure they understood what was going on so they were not just copying the code. Explaining line following while walking a line is one of my favorite activities.

This kids also spent a good deal of time experimenting on their own. I think the lessons provided a good foundation for those experiments. For example, the kids learned about sensor detection ( light, touch, US ) from the tutorial but they had to experiment on when to use a sensor wait block versus a one time sensor switch block. With this type of experimentation, the students quickly learned which types of operations force a wait in a running programs and which types require repeated calls via a loop block.


Related Posts Plugin for WordPress, Blogger...