May 31, 2009
The Mac can move in either manual mode (with a bluetooth remote control) or in autonomous mode, with the UltraSonic sensor used to avoid obstacles (and wayward R2D2 units!)
more info at thinkingbricks
May 30, 2009
May 29, 2009
Chris has a new book coming out in July titled "Free: The Future of a Radical Price" that builds on an excellent article he wrote for Wired a while back. While I'm looking forward to the book, I went back and re-read the article. I'd like to open up a wild discussion here and welcome your thoughts... so here it is:
One of the article's (and book's) basic ideas is that businesses might find growth opportunity by giving away certain products and selling others. Imagine if LEGO gave away a single NXT Brick to every science teacher who requested one - they couldn't do much with it without the software, motors, sensors, and other stuff, right? But imagine if they threw in two motors, two rubber wheels/hubs, a caster, two wires, and the software? Crazy to consider, but still, all you can do with it at this point is build a tribot of some sort with no sensors and very little functionality.
Could LEGO recover the costs of all those free NXTs (and parts) through the profits received by selling sensors, add-ons, etc? Would having the basic components of a programmable tribot open the eyes of those teachers unfamiliar with the NXT and encourage them to request the purchase of additional parts, lesson plans, and activities for the classroom? (Keep in mind that it can be argued that Apple had some success years ago by giving away an Apple computer to every school in the USA - the philosophy, I think, is to get them hooked on the Apple computer and they'll purchase an Apple later in life.)
Do I think this is realistic? Of course not... but anything's possible. MINDSTORMS has a well established community and teachers (at least here in the USA) seem to at least be familiar with it through events such as FLL, RoboCup Jr, and LEGO Education sales reps. LEGO is in business to make a profit through sales of their products, but it does make me wonder if giving away some of their key products such as the NXT brick might stimulate sales in other areas. (Imagine if every Technic kit they sold came with optional instructions for integrating an NXT brick into the design - not hard to do, in my opinion.)
from the website
This action sequence generates a demo dataset and renders the chart based on that data. At the Google I/O 2009 Sandbox, Pentaho demonstrated integrating an Inline ETL process with access to a Google Docs Spreadsheet.... whatever that means.
Nonethe less it looks cool and would make for a great school project.
It uses the LeJOS firmware and 2 NXT (presumably to allow for the 5 independent bars)
via Make Magazine blog
May 25, 2009
Two ways are shown to handle the steering. In the first method, the steering wheel on the remote control is used as an electrical generator to produce and send electrical power directly to the steering motor on the vehicle through a wire, without the NXT even being involved. In this mode, the steering even works when the NXT is off! In the second method, the steering wheel acts as a rotation sensor for the NXT, and the NXT then controls the vehicle's steering motor by program control through a different wire.
In the program-controlled steering method, the steering motor on the vehicle is acting like a "servo" motor, which seeks out the desired position as determined by the other motor (the steering wheel), which is acting like a rotation sensor. Doing this is surprisingly tricky, given that the end effect is so simple. The program measures the degree difference between the two motors and then continuously modifies the direction and power level of the servo motor to try to zero in on the desired position without any noticeable overshooting or oscillation (see the program provided).
Given the very simple construction of the rover (solid rear drive axle and parallel arm steering), turning very sharply doesn't work very well, so the model can also serve as a demonstration for why real cars need a differential and perhaps Ackermann steering. Some links are provided for users who want to try to investigate these topics.
Here is a video of the rover in action:
May 24, 2009
A new release of leJOS NXJ, the Java platform for the NXT, is out!
The new version 0.8 contains plenty of new features, bug fixes and improvements, including
- Controlling the NXT from the computer
- Faster bootup times
- Support for RS485
- Support for RFID, RCX Rotation Sensor, and EOPD sensors
- Support for Java 1.6 classes
- Support for Generics, Enum classes and foreach loops
- Support for subsumption architectures
- NXJControl utility for quick control of motors and sensors
- Wider support in the lejos.navigation packages
- javax.microedition.location for Bluetooth GPS
- Lots of sample code in the 'samples' directory
At the same time there will be a huge display of all sorts of miniature models and RC-models in Elst.
See for more (address) information: www.debouwsteen.com
May 20, 2009
Got a great email from Mariah S. along with some pictures...
"Hi Jim, Just thought I'd drop you a line and let you know what a great time our 4-H'ers had at the robotics camp this past summer. It was our first ever robotics camp and we decided to go with "The Mayan Adventure." 4-H'ers from across the State of Mississippi came together to see if thier robots could uncover the secrets of King Ixtua's tomb. The ExploroBot robot was a great 'first' robot for the 4-H'ers to build as some of them had never built or programmed a robot before. I thought that the robots and the challenges were structured in such a way that the youth were able to move through progressively more difficult challenges improving their skillset along the way. They especially enjoyed creating "StringBot" although we christened it "VineBot." All in all everyone had a great time. Thanks for writing such a good guidebook, can't wait to see what's next!"
Thanks, Mariah - love the pictures and glad the kids had fun!
May 18, 2009
May 17, 2009
"built a NXT robot that moves a digital cam around while it is shooting an image each second. The cam runs in multi shot mode on a firmware extension called CHDK with consistent light exposure. Once the robot has covered a 180° area, it moves down for 10° and continues. This way, a 180° x 60° panorama view image comes into being which is finally assembled on the computer. When making use of the maximal zoom factor of the cam, almost a Giga pixel of data is generated."
(Translation from the creators' statement)
Another movie with a panorama image of the Reformierte Kirche in Richterswil (CH) can be found here.
May 13, 2009
From guest blogger Nils:
I've built a robot that is hanging from the ceiling on top of a pile pf paper sheets drawing dots. It's part of an art and design exhibition in the Netherlands. The exhibition is about the way we look at things. While a visitor is looking at images of certain objects an eye-tracker records where exactly the viewer has just looked at. The Lego robot then reproduces exactly these points with a thick marker so you're able to see how objects have been perceived. By this way more and more large scale images are produced and shown on a big wall and it becomes comparable how certain objects have been seen differently. There are much more images, movies and information on my website:
May 12, 2009
May 9, 2009
May 8, 2009
Sorry for the delay with this post... I've been busy discovering just how much time it can take to prepare for college placement exams and Calculus AP tests. :-)
Here's a video that shows the capabilities of Power Surge's robot, "Spike". BTW, Spike is a reference to a voltage spike, which is another name for a power surge. The video shows what Spike can do in autonomous mode as well as tele-op (driver control) mode:
One of the robot components shown in the video that might not be as well known as other parts of the robot is our Intake System, or our mechanism for retrieving pucks off the floor. At the regionals, we had an intake system that could get one puck at a time, and that required quite a bit of accuracy on the part of the driver. Also, the puck had to be on a flat area. This made it slow and unreliable for us to get pucks off the floor in matches. In preparation for Atlanta, we wanted to improve our intake system, since we thought opponents would probably dump our pucks on the floor early in the match to try to prevent us from scoring them. We came up with an design - somewhat similar to a vacuum cleaner - that works really well. A rotating turbine with flexible brushes sweeps pucks into the robot. Then the same turbine rotates and lifts the puck onto vertical conveyor belts (made with LEGOs and non-slip pad), which lift the puck over 17 inches off the ground before dropping it into our big Puck Magazine with the rest of the pucks. The whole thing is done with just three NXT motors. It can retrieve pucks from almost anywhere on the mat: obstacle strips, up against field walls, etc. In addition, the mouth of the Intake System is 8 inches wide, so the driver doesn't need to be accurate at all; he can simply drive the robot over a puck in roughly the right place and the robot will take care of the rest.
Below are videos of the autonomous and driver-control modes of our 5th qualifying match. We're on the blue alliance, and our drivers have yellow shirts (one of them is me). We won this match 222 - 208. This was the highest individual score (222), as well as the highest combined score (430) at the World Championships. We used our autonomous mode that scored two puck racks for the first time in this match and it worked flawlessy, putting all 18 pucks into the triangular goal. It turned out to be a good thing that we did this, because a team on the opposing alliance also used a two puck rack-autonomous mode for their first time, and their partner also scored a couple pucks in autonomous mode.
Driver control was a little stressful for our fans. I was the primary driver, but neither the on-field coach nor myself had noticed that our opponents scored so much in autonomous mode. So in driver control period, we focused on scoring lots of our opponents' pucks to obtain ranking points and overtake the #1 team in the rankings (see my previous post about FTC). However, it ended up being better for us, because we still won the match, plus we got a lot of ranking points and took over the #1 spot.
You can see more information and pictures from the World Championships on my team's website: http://www.teampowersurge.org/
May 6, 2009
I put them into my brick that I'm using for testing robots in a new book (where I'm constantly stopping, starting, stopping, etc.) and decided to see for myself. Having used a mixture of alkaline and rechargeables in the past, I have to say these things are unreal... I'm still using them after about 2 weeks of daily use and they're still showing 8.X on the battery meter. I've seen no change in performance either.
They're a little higher priced than the batteries I'm used to using, but for testing purposes, I'm really leaning towards switching over to these during those critical testing periods. It made me think of FLL teams who are also constantly testing and need reliable batteries.
I'm interested to hear from our readers on their battery experiences. For those who don't use the NXT rechargeable, what kinds of times are you finding on the batteries you use? What do you use?
You can also view a TV interview with Daniele following this link. The video also covers some of his other creations.
May 5, 2009
One of my favorites: the "Ultimate 100 Meters Race", including an ingenious analogue multi output device:
Have a look there - it's worthwhile!