Monday, March 30, 2009

Aerial Laser Tag, Continued

... Continued from Aerial Laser Tag...

Since we have no direct synchronization b/w the blinking of the laser and the sampling of the light intensity, the quadrature aperture sampling method ensures that we can achieve this effect even though the frequency is slightly off. the end result is that we are able to detect when a modulated laser beam is shone on the light sensor whereas any other light source will be ignored.

THe laser 'gun' is made by using a microcontroller to modulate the laser beam, which is then turned on by a switch, (a trigger) which when pressed, sends out a modulated light beam for fifty milliseconds and after that, pauses for 200 milliseconds thus, only allowing firing of the gun approximately five times a second. The beam is modulated at 20 khz and the demodulation sampling techniques allows the intensity of modulated light to be registered 40 times per second.

The receiver consists of a microcontroller with a light sensor, two LEDs, and a plastic cup! The plastic cup is placed over the light sensor, diffusing all incoming light into the light sensor. Since the light sensor is so tiny, the diffuser is needed to ensure that light can reach the light sensor. Thus, when the laser is fired at the receiver, its light is diffused, and since the laser's frequency matches the desired frequency, it is accepted, and registered, turning on the LEDs to indicate a 'hit'.

We've tested pointing four other laser pointers individually and simultaneously at the receiver and only when the correct laser with the desired frequency is pointed at it, does it register and give feedback via LEDs. We have also tried using the system in direct sunlight when the laser is barely visible and it still registered, even at a distance.

So what does this mean? It means that the receiver is ignoring all other light than the desired frequency. Back to our story of the plane game, this is our prototype to see if the technique works for having an outdoor transmitter/receiver that cannot be disturbed by other light sources such as ambient light. The aim is to make multiple transmitters and receivers that can be worn by people to create a two way laser tag game between people on the ground and the plane firing lasers from overhead, and people returning shots to the plane from below. Schematics available soon, so you can build it yourself!

More photos here:

Movie Here:

Thursday, February 19, 2009

Fun with Building Robots

GeekPhysical and Illutron did a workshop this week in Odense, teaching students all about Arduino, electronics, physical computing, using sensors, and building robots from RC Motors! We had a ton of fun, and were happy to see people being creative with their robot building.

Our robots consisted of servo motors, one small and one big, glued to each other with the smaller on top. This one had a stick glued to it which could be used to pick up objects. Participants were taught using Pure Data and the pduino interface so they could easily associate the programming with what they were doing.

Our next goal is to build a patch that allows sensors AND the servo to be connected at the same time, so that we can use sensors to control the servo! In the meantime, check out to see what the next day, and a couple of guys crazy about computer vision used the robots for. Hint, control a robot with fruit! Woohoo!

Video on bliptv here:

Friday, November 7, 2008

Biometric Jewelry Continued...

The circuit board has been routed using a CNC machine to create the intricate details, twists and turns that make this piece unique. In case you're not familiar with how circuit boards are made, here it is. A very strict and detailed process was followed to create the project. First, a design was created using an image of a celtic triangle. This was then mapped into a vector art program, and traced over. The circuit diagram was laid overtop, and all cut lines, drill points and cut outs were defined. To create the circuit board, first transparencies were printed, matched, and is contact copied in a UV light box which transfers the image from the transparency to the circuit board. The circuit board, which is covered by photo resistive material is then covered with a chemical, which removes all the photo resistive material except that which has the pattern from the transparencies imprinted on it. The circuit board is then put in an etching machine, removing all the copper not covered by the photo resistive material. This reveals the copper tracks.

The CNC machine is used to drill holes where they are needed (for components, etc). The outlines of the pattern are then routed into the circuit board and the outline of the shape is established. The CNC machine then details the shape, drilling and etching to reveal the final celtic knot triangle.

Once out of the CNC the components were soldered on to it and all copper traces are covered with solder. Components that should not be painted, such as the LEDs, are masked off with masking tape and then is sprayed with paint and put in an oven to cure. The tape is removed, and a glowing heart rate monitor is revealed.


The LEDs are programmed in a special way since the micro controller (AVR Tiny45) is so small, with few pins, all the LEDs are connected in such a way that only a single one can be lit at a time. In order to give the impression of each LED having individual brightness to create the waving motion, very fast multiplexing was implemented. (i.e. switching between LEDs very rapidly, varying the time each LED is turned on).

The necklace itself has a heart rate receiver, receiving pulses from the heart rate monitor.

The heart rate portion is borrowed from a previous project, so every time a heart beat occurs, the LEDs light up in a pattern wherein each LED receives an individual brightness to simulate ambers glowing in synch with the heart rate, and fading out over time.

Physical Hardware:

The processor, five LEDs and a capacitor were all that were used on the necklace. In a moment of minty inspiration, we used a Listerine breath strips container to house a lithium polymer battery, however we are contemplating using lithium coin cells instead for easier replacement rather than charging.

Wednesday, November 5, 2008

VFD (Vacuum Florescent Display) Clock

After a certain someone nearly threw our 'alarm clock' (mobile phone) out the window the last time Britney Spears woke us up (Blackout has GREAT wake up songs) it was decided that perhaps we should have our own alarm clock. Since a friend inherited 15,000 of said VFD tubes (Russian Made), and had some odds and ends from an electronics factory that were no longer being used; we decided it would be interesting to solve our 'alarm clock problem' with these.

Components are new but have been rejected from the assembly line in the electronics factory, normally due to bent pins and other mild deficiencies. We saved these throw-aways and used them for the clock. The circuit boards used were custom designed and manufactured using a rapid prototyping process (CNC Machining). Three plates (boards) were specifically designed; one for holding the displays and interface button, one for the clock electronics and one for the speaker, and to act as a base for the entire clock.


The processor uses a 16mhz crystal to generate the time signal. The 16mhz signal is first divided by eight and then by two thousand to generate a one second pulse. The program counts the seconds to form the minutes and hours. These are then displayed on the tube.


The interface was designed to give the user a fluid experience when setting time, alarm, and snooze features. The time is set by holding down button one, which stops the dot flashing. Once the dot stops flashing, the time can be programmed. Hours are programmed with the left button (1) and minutes with the right button (2). To set the alarm, the same process occurs but initiated by the right button (2). When the alarm goes off in the morning, there is a snooze feature, which can be 'snoozed' by pressing left (1). To turn off the alarm, hold down the right (2) and press left (1). The best feature? An innovative snooze twist, the clock flashes between time display and the word "Sn" when it is snoozing! Finally you know if your clock is on snooze or if its turned off... We are working on a new feature which may display how many minutes are left between the last snooze and the upcoming buzz. Speaking off, the buzz is hopefully going to be replaced by our great Arduino synth which is capable of outputting some truly fantastic music. We are also considering a dimmer function, the VFD tubes are capable of having their brightness adjusted, and the circuits are prepared for a light sensor, we just have to work it into the interface!


The electronics for driving the VFD tubes is a bit tricky, it needs a slightly higher voltage than the normal 5V signal used for typical LED clocks. The VFD uses 35V as well as a 1.2V filament voltage. Apart from driving the clock and user interface, the processor also has to control the generation of these voltages. Each VFD tube works like a regular 7 segment display, and are actually driven exactly the same way. The processor switches rapidly between the four tubes, so that only one tube is lit at a time (multiplexing). As such you can see eleven little legs, (pins) eight of them control the anodes (segments) within the tube. Two of them have filaments connected between them, and one has a grid which enables or disables the entire tube. All the anodes for the four tubes are connected in parallel and the processor uses the screen to switch rapidly between the tubes. This makes it possible to write individual digits on each tube using only eight common pins.

Wednesday, September 17, 2008

Active Energy Harvester

We are investigating different survival techneques for micropower devices. Rather than cleverly managing whatever energy is available to an inanimate device it is interesting to see what advantage devices that actively hunt for energy may have.

This project is a solar powered robot that uses a tiny RC-servo to slowly scan the surroundings for solar power. Like some flowers it follows the movement of the sun during the day.

All energy needed to move the solar panels and run the robots microcontroller is stored in an tiny Lithium battery during time with excess solar energy. During nighttime the mechanical function is shut off and only the robots brain is kept running. During periods of energy starvation the robot goes into deep hibernation saving the battery from being destroyed by deep discharge.