You can download it from Steve’s page here, which has some other cool blocks and many good NXT stuff.

In a few days I will post instructions on making new fonts. I will happily add contributed fonts to future releases of the block. Also, I learned a few new tricks on block building which I will share in the near future… Comments/questions are welcome!

Download the .zip package here

Guy Ziv

Nice work!

Stefan Vorst designed the UCB, a controller small enough to fit in a 2×4 Lego brick (the picture above shows a UCB that is not housed in a brick). It comes with a power supply that connects to a Lego 9V battery pack; the power supply fits in a 2×2 brick (shown in the picture). The UCB can control two Lego motors. It also has a 12-pin connector that can be used to control up to 6 LEDs, or to interface to sensors. An expansion connector can link the UCB to an NXT; the NXT can then control the UCB. In this mode, the UCB can function as a motor controller, LED controller, or a sensor unit. The UCB is programmable; the device in the background  is a Pickit 2, a low-cost microcontroller programmer that can control the UCB.

To learm more and to see the UCB in action, check out a video that shows it controlling motors, connected to the NXT, and being re-programmed.

The UCB has been adopted by Stefan’s club, the LEGO^® Modellbaufans Rheinland e. V. Code for the UCB is available on sourceforge.

The details, movies and more explenations are here:
http://www.teamhassenplug.org/sumo/PSumo.html

Here is our new sound localizing robot called Nic_3. It is capable of fixing the azimuth of a sound source that is located in the horizontal plane. The localizing process is done using the stereo audio amplifier that we already used with Nic_1 and Nic_2. (For several reasons Nic_2 has not been published yet.) The amplifier is directly connected to the RCX inputs 2 and 3. (input 1 is used to power the sensor).

The RCX samples 2 x 1000 data-points on each ear at a frequency of 36kHz. Then it processes the time-lag between both signals within 100ms. This is impressively fast, because we apply a high speed phase detection.

The RCX repeats the sampling and detection process 10 times and takes the average of the measurements. Then the RCX computes the azimuth from the time-lag and sends the angle to the NXT through the HiTechnic IR-link.

The RCX program has been written in ULTIMATE ROBOLAB and H8-Assembler, while the NXT has been programmed using the LabVIEW NXT-toolkit.

http://www.convict.lu/htm/rob/phase_detection.htm

Here’re details from NXTLiftBoy:

This is my second attempt at a forklift from a single NXT set.

On the model:

I think this one has better looks and a better design than the first one. Getting pictures when taking it apart is planned for, like an upload to photobucket so anyone with a NXT and a differential can build it afterwards.

The earlier forklift had a caster wheel that I wanted to avoid this time, and I wanted the center of gravity down for better stability. I had to resort to allow myself a single part that is not in the NXT set, which is a differential.

I started with the steering back wheel construction that has a very tight fit. Then I connected the second motor, but I failed trying to gear it onto the differential, so I took the third motor and mounted it standing up, which worked out better, because the motor axis was now parallel to the differential. Later I fortified the lose end of the motor gear axis, to prevent gears slipping when fully accelerating in a turn.

The NXT brick was mounted just on top, but it had to be moved to the front as far as it could to prevent the inner front wheel from slipping when taking curves. The brick has the USB socket backwards to allow for easy “fuelling” of software!

Under the brick a cable duct is formed by two beams. The motor cables run from the back through the duct to the front and bend in rather tight turns, due to very compact building. One has to build it with the cables inserted in the motors, wiring them afterwards is totally impossible.

Then I made a mounting point for the mast, and fortified the construction to deliver the forces when lifting onto the front wheels.

The mast is the most obvious part I guess. I optimised the fork design to fit IKEA boxes. These boxes are the same as those colorful used for my NXT Container Crane. Routing the rope was not so easy, it takes a turn in the end, because the winch has a vertical axis.

From the remaining parts the cabin was made. It features a roll-over cage, backside motor lid with air inlets, chair with captain left armrest with controls, rightside controls, steer with turning knob, dashboard, gas- brake- and clutch pedals, left outside mirror (tilts and swivels), 4 outside spots for lighting the work place, backlights.

Finally the US transducer was put in. I first had it ‘looking’ sideways. I intended to program it to count the rows of IKEA boxes it is passing, take a left turn at row X, then count the columns it is passing take a left turn at column Y, in order to fetch one or more stacked IKEA boxes at height Z! That would allow for a fully automated warehouse.
I was skeptical when it comes to the required precision… and indeed i did not manage to program it properly, as without additional sensors (compass!) tiny errors accumulate.

The US sensor is now on top, so it can in principle see how many boxes it is approaching…

To have the forklift doing something very useful (ahem) and funny, I came up with the idea to have it solve the Towers of Hanoi:

Towers of Hanoi is a well-known puzzle with three vertical pins. It starts with a pile of discs with top-to-down increasing diameters slided onto the first pin. The task is to get the pile to the third pin, but only a single disc may be moved at the time, and a bigger disc may never be put on a smaller disc. This is typically solved using a programming technique called recursion. For 4 discs already 15 moves are required.

Rather than pins and discs, i wanted to start with a pile of IKEA boxes numbered top-down 1, 2, 3. A higher-numbered box may never be put on a lower numbered box. The forklift is programmed to move the stack of boxes, one box at the time according to the rules, from the starting position on its left side to the end position on its right side, while only using a third centered position for intermediate storage.

I failed to obtain sufficient reproducibility in moving, trying both NXT-G (the LEGO language) as well as Java iCommand. Even preventing slipping by gradually adapting speed would not help. Maybe LeJOS can do the trick?

So I challenge all to show me it can be done (without additional sensors).
I guess programming is not my strongest quality

Happy lifting once more,

NXTLiftBoy

Josh

“Speakers for your iPod can be clunky affairs. They very rarely fit in your pocket, and we’re almost certain none of them look like a building block, even if they’re almost the same size. That is until now! The iBlocks have been cleverly modelled on one of our favourite childhood building blocks (you know the ones we mean) and are small enough to pop in your pocket, so there’s no need for extra cases or wires to clutter up your bag. The sound is pretty impressive for something so small, the Active Bass System makes sure the speakers give out quite a kick. All you need to do to get it working is to plug it in and we just love the volume control buttons that have been camouflaged to look like ‘bumps’ and discreetly placed in the middle of the block. Compatible with 3rd Generation iPods onwards and the iPod Touch.”

They come in red, white, and black.  You can buy them here for £12.99

Josh