Glowworm Love

What is it?

“Glowworm Love” is an ambient room light, consisting of 4 individually colored artificial glowworms. These worms can be combined by bringing two glowworm bodies together. If they like each other, they will mix their color and produce a new color. They will keep this new color even after splitting them again.

This way the user can produce many different colors and create different moods within the room by playfully combining the glowworm lights.

Bionic principle

While searching for interesting phenomenons and mechanisms in nature we took a closer look at glowworms (like the ones that can be found inside caves in Australia or New Zealand) and at the way they use the glowing light which they can produce.

The purpose of the glow varies. Some worms are believed to glow as a warning signal to predators not to eat them as they are mildly toxic. Others glow to attract prey. But we found the most interesting usage of the glow to be communication: adult females that glow do so to attract a male for mating.

And with that principle of using light to find and attract the right partners the idea of “Glowworm Love” was born.

Adapting and adjusting the idea

So we decided to build artificial glowworms. There would be several worms and they would use light in some way to interact with and react to each other. Using different colors for the glowworms would then create some very nice effects and would allow for all kinds of atmospheres to be created within the room.

But how should our glowworms interact exactly?
The idea that came to our mind was to simulate the mating of two glowworms. There should be different colors and/or light signals that would indicate matching and non-matching combinations of worms so that only the “right” partners could come together and “fall in love” with each other.

Of course we would not be able to make our artificial glowworms produce little glowworm children so we thought about something else: what if matching glowworms would – instead of producing new glowworms – give birth to a new color of light? This idea was very appealing to us and we decided to go with it.

Color comes into play

To step it all up a little bit we wanted to provide the user with some kind of challenge and generate some learning effect: to do so we derived the rules for matching glowworms from the RGB color model.

This is an additive color model in which red, green, and blue light is added together in various ways to reproduce all sorts of different colors. The name of the model comes from the initials of the three primary colors, red, green, and blue. The RGB color model is used in all sorts of electronic display devices where red, green and blue pixels are combined to all other colors as follows:

the 3 primary colors can be displayed simply by lighting the corresponding red, green or blue pixel. Yellow is produced by combining red and green, cyan by combining green and blue and magenta by combining red and blue. Finally, white is generated by lighting red, green and blue at the same time. Black – which means no light at all – is simply achieved by turning all pixels off.

By gradually mixing all the aforementioned colors it is possible to create all the colors the human eye can perceive.

For our glowworms we decided to stick with the pure primary and secondary colors red, green, blue, yellow, cyan and magenta without gradually mixing them. This makes it possible to have the following rule-set for the combination of worms:
Primary colors can be mixed to get secondary colors. Secondary colors can be mixed back to primary colors again. So red, green and blue worms can be combined. They then both change their color and turn into the secondary color given by the (primary) color-circle.
Furthermore yellow, cyan and magenta worms can be mixed and then both turn into the according primary color, again given by the (secondary) color-circle.
The following graphic illustrates that process for two exemplary combinations:

This implies, that for example a red and yellow worm can not be mixed, because red is a primary color and yellow a secondary (in fact, the mixture of red and green). When mixing two worms of the same color, nothing happens and they do not change.

These rules can eventually lead to configurations, where no matching worms are left to combine. One such combination would be one red worm and three yellow worms in a system of 4 worms in total (worms of the same color can not be mixed, red and yellow can not be mixed either). To solve this problem we needed some sort of mechanism to release the system once it got stuck. So we planned to integrate a reset button to each worm that would change its color back to the original color it had after turning the system on.

One thing we wanted to do as well was adding the possibility to change the brightness of each worm independently to adapt our glowworm lights to different situations – one could make very bright worm lights or dim the worms down to create a more moody atmosphere.

Our final concept

Finally our conceptual system contained:

• 4 individual artificial glowworms
• RGB LEDs within each worm to control its color
• the ability to connect 2 worms and mix their color in respect to the mixing rules
• a reset button on each worm to change it back to its original color
• a potentiometer on each worm to control the brightness of its LEDs
• a micro-controller to run the mixing algorithm and control the LED colors

Building the Prototype

After deciding, what elements and functions we wanted to integrate into our prototype we had to think about how to realize and design the glowworms.

We found that clear plastic spheres would be a pretty asthetic and very practical solution for the glowworm-bodies. A few spheres of this kind attached to each other in a row then form a worm. All technical stuff like LEDs, cables and switches can then be put inside the plastic balls.

This way we got a nice, modular structure that made the process of building all 4 worms quite effective: first we prepared all spheres and built the LEDs and switches separately and then we put everything together in the end.
This straightforward “mass production” was crucial due to the limited amount of time we had to build the prototype.

The different steps of building the prototype were as follows:

Writing the code for lighting and mixing the RGB LEDs
This was done within the Arduino coding environment. We tested the algorithms by simply connecting some LEDs to the Arduino board directly.

Making the LED stripes for the individual glowworms
As we had no ready-to-use LED stripes in a length that would have fit our design, we had to adapt the RGB LEDs by cutting and soldering them back together as well as adding wires to them.

Making the worm-bodies out of 25 plastic spheres
After we got simple decorative plastic balls from the handicraft store we had to cut holes into the spheres to create places where the LED stripes and all cables could go into and out of every sphere. To cover the inside and make the light emitted by the LEDs more ambient and soft we scratched and sanded the surface of every sphere until it got a white, milky look instead of being fully transparent.

Designing and building the mix-switches
This task was more complicated than expected. Since the system needs to be able to detect which two worms were put together, we needed some mechanism for that.

Due to our limited resources and building material we came up with the following plan:

This 3-point-switch controls 3 seperate electricity circuits. As each circuit can be open or closed we have effectively a 3 Bit encoding mechanism that allows us to distinguish between 8 different states, each one being a combination of open and closed electricity circuits. Since there are 6 different possibilities to connect the 4 glowworms this is a sufficient number of states to clearly differentiate between all possible combinations. We cut the switches out of plastic with a laser cutter and attached 3 metal plates to each one. These plates resemble the contacts for each of the 3 electricity circuits.

Designing the reset buttons
The reset buttons are simple on/off-switches. When the user presses the button, the correspondent worm will reset its color. We found it an interesting and funny idea, to put the buttons on top of some “eye” of each worm. This way the user feels like “I’m squeezing the glowworm’s eye” instead of “I’m pressing a button” which makes for a much more natural and vivid experience. At the end of the day the user shall play around with the glorwworms without thinking too technically.

Enabling dimming of each worm
To integrate the possibility to change the brightness of every worm, we made the RGB LED’s color values dependant on a potentiometer attached to each worm-body. By turning the potentiometer one can change from full brightness to complete darkness and everything in between, since it is an analogous input. Again, we wanted to integrate the potentiometer well into the glowworm-metaphor, so we decided to attach it at the end of the worm-body to represent some sort of tail.

Putting everything together
Finally we had to combine all 4 individual worms by connecting them to the Arduino board and hiding all electric cables and bread boards inside a cardboard box. This way it became possible as well to attach the box to the ceiling and make the glowworms hang down from it.

Values and potentials

“Glowworm Love” can obviously be used as a fun and eyecatching ambient room light. The number of worms is not restricted to 4. Extending the system would allow for use cases and scenarios only limited by imagination, ranging from small systems with 4 to 6 worms within private rooms (living room, bedroom etc.) to large applications in restaurants or even public places with hundreds of glowworm lights.

Another value of the system is the implied rule set. Mixing light colors is something people are not used to from their everyday life. Everyone has learned how to mix pigment colors like crayons or watercolors, where red and blue create purple, blue and yellow create green and so on. But the RGB color model is different and more abstract and therefore not that intuitive. Surprisingly we realized that playing around with “Glowworm Love” changed that totally. After a short time we felt that mixing RGB colors was as natural as mixing pigment colors. So there is a great potential in the system to be used as an educational device to illustrate the RGB color model at schools, universities or museums. The fact that “Glowworm Love” is quite an eye-catcher would draw people to it and encourage them to experiment and learn.

Next Steps

Having build the prototype the upcoming steps would include testing the system in different scenarios with different users and improving and extending it. One thing that should be improved is the detection of worms being mixed. Our 3-point switch was sufficient for this prototype, but another solution where worms could be combined at any point of their body would be nice and worth striving for. Another thing would be reducing the size of the system. Hence this is a rather rudimentary prototype, everything was build from quite huge components and connected via breadboards. Printing circuit boards and using smaller components would reduce the size down from the big cardboard box to a cigarette-box-sized system or even smaller. It would be interesting as well to experiment with different shapes and sizes for the glowworm bodies. Very small glowworms could be created as well as very large ones. There are many more things that could follow building this first prototyp like adding more functionalities to the software to create even more effects or implementing other rule sets for mixing the glowworm lights.

Problems we encountered and tips for upcoming participants of “Sketching with Hardware”

RGB LEDs
If you plan to use LED RGB strips as we did within your work, have a look at the tutorial that the following link points to:

http://www.ladyada.net/products/rgbledstrip/

And do not forget, that you need 3 transistors for each RGB LED! So make sure there are enough there for your work…

Arduno Uno vs. Arduino Mega
We started with an Arduino Uno Board just to realize rather soon that there were not enough pins at all. We needed 12 pins for the RGB LEDs, 3 pins for the reset-buttons and another 3 pins for the 3-point mix-button. And finally we needed 4 analog-input pins for the potentiometers. So do not underestimate the number of pins you need, especially when working with RGB LEDs!

Blowfish – Don’t scare it, or it puffs up!

What is it?

This years theme in “Sketching with Hardware” was Bionics. Consequently, the task was to rebuild or reuse a natural and biological method. Our team decided to imitate the behavior of Tetraodontidae – that’s the family of fish with the ability to inflate their body to a ball-like shape in order to defend itself. In an outburst of pure creativity, we named our prototype Blowfish!

Blowfish is designed to puff up when users come too close or are too loud. Somebody clapping hands loudly will find Blowfish doubling its surface. As puffers don’t like people coming too close, Blowfish inflates if one approaches it’s face.

To put it in a nutshell, Blowfish is an ambient display visualizing noise and distance in a striking way. Just put it in one of your room’s corner and you’ll always know whether or not your shouts about an exiting soccer game will disturb your neighbors.

[ Skip to Video-Demonstration ]

How does it work?

The basic building blocks of the prototype are:

• umbrella
• printer slide
• metallic arm that pushes the umbrella
• relay, switching the motor direction
• front emergency button to switch the direction
• clap sensor
• distance sensor
• microphone in combination with processing
• serial connection → Arduino → control motor power

The basic concept is to push the umbrella open, by moving the carriage of a printer slide.

In order to extend the reach, a metallic arm is mounted on the moving carriage.

In the same way a real blowfish deflates it’s body after a certain amount of time, our blowfish has to reduce it’s surface area. This is achieved by moving the printer slide backwards which in turn drags the umbrella shut. On a more technical level, this means that the motor powering the printer slide has to reverse it’s direction. The exact components of how this was achieved are detailed in the next section Lessons learned.

As the motor power is controlled via an Arduino, it was possible to install an emergency button at the front end of the printer slide that notifies the Arduino each time the umbrella is opened completely.

Animation showing the movement of the printer slide. Note: This is not the way the carriage moves in the final blowfish prototype.

After a fixed amount of delay, the Arduino switches the motor direction and gives power to the motor in order to close the umbrella.

As blowfish is sensitive to noise, the prototype requires some kind of sensory input to recognize sound volume.
The Arduino is connected via USB to a laptop which functions as the power-source. In addition to providing current, this connection can also be used to control the Arduino’s program logic from software, for example by using Processing.
The principle of operation is as follows:

Microphone input → Processing → check threshold → notify Arduino via Serial (USB) Connection → Arduino initiates ‘alarm function’ (starts motor) → umbrella opens

In addition to volume sensitivity, blowfish also react to physical distance. This ability is reproduced using an IR sensor which is directly connected to the Arduino. The Arduino’s software computes median values from the sensor’s raw input in order to dampen the signal’s oscillation. Finally, a threshold check is performed and the ‘alarm function’ called when the signal is above the set threshold.

Lessons learned

How to switch the direction of a DC motor?

As stated above it was crucial for the blowfish project that the device is capable of closing the umbrella, in order to deflate the blowfish body. That means that the motor of the printer slide has to be able to move forward AND backward. This can easily be achieved by consecutively supplying positive and negative voltage to the DC motor. Theory is often so easy … while bringing this concept to life in a working prototype is not, especially not for non Electrical Engineering students.

After further research, we discovered that it’s possible to use an H-Bridge to apply voltage across a load in either direction. We implemented a circuit using relays to allow the DC motor to run forward and backward. In fact, two relays are used. One to stop the motor and one to switch the current flow.

Use 2 axicom d2n relays to turn on/off and switch the direction of a DC motor.
Pin 9 controls the running direction of the motor while pin 8 turns it on or off.

The basic idea is to use a DPDT (double pole double throw) relay which separates two differently polarized circuits. Without activating the relay, the motor is +/- connected. When voltage is applied to the relay (the switch inside changes its position) the motor gets -/+ connected and runs in the opposite direction.

Oil and measuring matters

Designing the lever or arm pushing the umbrella open or pulling it shut guided us to the field of solid handcraft. Metal bars needed to be cut, bended and drilled. We soon found that accurate measuring simplifies mounting and unmounting of components. But if it comes to grinding metal devices due to not perfectly matching screw holes oil helps you out.

Parallelize your work

As we were a team of three students with a rather challenging project, it was necessary to split the work between each of us. We had to carefully think about which step could be done at what time. Team organization included classifying the steps into mechanic, electronic and informatic work. Thus each of us could work at the same time at a different desktop. For example one team member implemented the clap sensor, while another one was soldering the electrical contacts and the third one designed and cut the materials for the final casing.

Values and potentials

The most apparent use case for our prototype is it being installed as an ambient display.
Using it’s distance and noise sensors, the blowfish reacts to noise by inflating and deflating it’s volume.
Congruent to this use case, it could be employed as a funny danger sign, expressing:

Don’t come too close!
(near a freshly painted wall, a wet floor, a hot surface …)

Next steps

To further increase the social component of the protoype, we plan to implement intermediate steps of inflation. This way, the blowfish can express his mood in a more differentiated way.
These intermediate steps of inflation could also be used to simulate breathing behavior.
This feature would dramatically increase the prototype’s impression of being alive.
Another attachment which will possibly be added to the prototype are movable fins. These would enable further forms of social interaction, for example expressing a ‘greeting gesture’ by shaking the blowfish’s fin.

Video

And finally: A video summing up the development process as well as the final presentation of the blowfish!

Some more photos showing the final installation:

FrettyFlytrap

Michael Konrad and Clara Lüling

lueling@cip.ifi.lmu.de

What is FrettyFlytrap?

Natural model for FrettyFlytrap is the venus flytrap, a carnivorous plant that is domiciled in a narrowly restricted spread-area in the USA. It catches and digests animal prey – mostly insects. Its trapping structure is formed by the terminal portion of each of the plant’s leaves and is triggered by tiny hairs on their inner surfaces. When an insect or spider crawling along the leaves contacts a hair, the trap closes if a different hair is contacted within twenty seconds of the first strike.
[Source: Wikipedia]

Natural model: Venus Flytrap
(Source: stock.xchng)

In contrast to the real flytrap, FrettyFlytrap isn’t just a plant that has primitive needs like food, but also has some human-like feelings: it seeks for love and fondness, and if it gets no food or if it is provoked, it probably shows its angry side.

How can FrettyFlytrap express its feelings? For the one thing, LEDs on the top of the plant’s head show its mood. Red LEDs signalize a critical state, green LEDs show that the plant is satisfied. A further indication is given by the plant’s mouth: If it is open, the plant is hungry and thus impatient or was recently provoked by being touched at its teeth. If you want to treat FrettyFlytrap well, you can tickle it under the chin, and you will see that FrettyFlytrap rotates its leaf, which is a sure sign that it enjoys your treatment.

How does FrettyFlytrap work?

The most important question was how to let the plant’s mouth snap. After having tested two different kinds of mechanical construction, it was decided to install a power servo motor inside the plant’s pot (see figure). The motor rotates an arm by 180 degrees. At the end of this arm, a robust but flexible stick is fixed and moved up and down. This movement is transferred to the plant’s upper jaw. A further motor was fixed at the stem of the plant to rotate a lightweight paper-leaf.

Mechanical construction: power servo motor

After the mechanical issues had been solved, the plant needed to get some senses for the users’ inputs. In this case capacitive sensors were most convenient. The plant’s teeth were painted with an electrically conductive silver-containing coat. For making the plant’s chin sensitive, graphite spray was used because its brown color is more suitable for the plant’s look.

For getting informed that a user had fed the plant, a push-button was installed at the bottom of the plant’s stem. If some food is thrown into the mouth, it falls down the stem and pushes the button which gives a signal to the arduino.

To breath life into the prototype, the last step was to deal with the program logic. Heart of our prototype is an Arduino Uno board, which contains the program code and controls the prototype. The program code declares an interval for the points when the plant gets hungry, for example every three minutes. If the Arduino board is informed by the capacitive senses about some user input, it introduces appropriate outputs (LEDs blinking/lighting or motor rotation).

Values and Potentials

The intension of FrettyFlytrap was mainly to create an entertaining toy. But it can be more. Despite it is just a simple electronic device, it seems for the user that it has a human or animal character. Thus it is imaginable that the prototype could serve as an amusing electronic pet the user has to care about. The use is easy and needs no explanation. The users have to find out how to treat the plant and are excited about its reactions. They can enjoy if the plant shows its nice side, and are maybe frightened if it bites them. Besides, in situations of no current interaction, FrettyFlytrap is just a normal plant – a decorative object.

Next steps

The current version of FrettyFlytrap allows a very limited interaction yet. But there are much more in- and output possibilities, for example sound. Extending the plant’s interaction possibilities would give it a more complex character and thus would make it more interesting.

[Like a Bird]

Marion Koelle and Marius Hoggenmüller

{lastname}@cip.ifi.lmu.de

The bionic paradigm?

Birds have a sophisticated technique to survive the frigid temperatures of winter. They possess a dense coat of feathers which can be puffed out to trap little pockets of air close to the bird’s body. These airpockets insulate the bird’s body and protect it from excessive heat loss. With this technology birds can sustain their body temperature of approx. 40° C even at extreme frost! At chilly temperatures they therefore often look like small fuzzy featherballs (see figure 1).

Figure 1: Sparrow with puffed up feathers

[Daniel Lingehöhl: Vogelwelt im Wandel – Trends und Perspektiven. 2010. Wiley VCH Verlag GmbH.]

[Image source: http://www.sxc.hu]

What is it?

For our project we adopted the birds’ anti-freezing technique and integrated it in an environment-aware interactive headpiece – at first glance: a simple, every-day bobble hat.

The basic principle is very simple: if the ambient temperature drops below a certain value or the user feels cold the heat-insulating functionality of the hat is activated and the ‘feathers’ of the hat are puffed out. The hat’s control system involves two modes, one based on the sensing of the ambient temperature and another one which is based on user-input via a touch-slider located at the hat’s brim .  It is possible to switch between modes by turning the hat’s bobble.

How does it work?

Build-up

The skeletal structure of the headpiece consists of two rows of inflatable ‘feathers’ made from a customary air mattress, which are comprised by a cover made from felt  (see figure 2).  If the compressor is turned on the air takes its way through a system of flexible plastic tubes located at the hat’s back and bloats the inflatable pieces which causes the hat to puff up.

Figure 2: Inflatable 'feathers' and felt cover

Pneumatics

To provide the barometric pressure which is required to inflate the hat, a small (air brush) compressor is used. As the compressor’s internal control can’t be easily accessed due to its direct connection to line voltage, a workaround had to be created. The toggle switch which turns the compressor on respectively off is operated by a servo motor (modelcraft RS-2) equipped with a 3-sided screw.

Figure 3: Servo motor attached to the compressor's toggle switch

Arduino’s Servo Library allows to determine a maximum and a minimum tilt angle between which the servo motor alternates. The compressor’s air duct is affilated to the system of plastic tubes that bloats the inflatable parts when the compressor is turned on. Once turned off, the elusion of the air is induced by the self-weight of the inflatable regions.

Tips & Tricks: Pneumatics

• Reduce the length of the plastic tubes where possible: unnecessary long connections cause loss of air pressure!

• Make sure all points of contact are airtight: tape is good for testing as it is removable, but it’s sealing is generally less tight than a permanent bond e.g. with hot adhesive or hot glue.

Sensor System

The system involves two different types of sensors: a temperature sensor and an array of capacitive foils that form the touch-slider which can be controlled by user input.

To determine the ambient temperature a LilyPad Sensor MCP9701A, which is a small thermistor type temperature sensor, is used. The analog output of this sensor is 19,8 mV per degree. The conversion of the sensor output to degrees depending on the input voltage (3,3V  or 5,5V) and model is described in [Post 4]. If the temperature obtained from the sensor drops below a certain threshold – for the presentation we simulated a sudden decrease of temperature with customary ice spray – the compressor is activated and the hat puffs up.

Figure 4: LiliPad Sensor MCP9701A and ice spray can

To enable the user to control the hat manually, according to his individual perception of temperature, a capacitive touch-slider is fit to the hat’s brim. The advantage of capacitive input is that only low to no force is required for the sensing, which makes it comfortable to control. The touch-slider consists of an array of four capacitive foils. In our prototype we used stripes of aluminium foil, for further analogical use-cases it is also possible to use copper strips or conductive laquer. For the implementation on the Arduino’s side we use the Capsense Sensing Library. The electrical capacitance of the human body can be sensed, when the aluminium stripe is touched with the finger. Therefore each stripe is connected to GND (ground), a high value resistor (1 MΩ),  a capacitor (10 pF) and one of the Arduino’s digital pins. The capacitor is not mandatory, but it improves the stability and repeatability of the obtained values. The exact connection diagram is shown in figure 5.

Figure 5: Circuit layout

Figure 6: Touch-slider

Tips & Tricks: Sensors

• If the data retrieved from the sensor is jittering you can anticipate this effect by inserting a capacitor to smooth the output values. An example of such a circuit is shown in figure 7.
  

  Figure 7: How to use a capacitor to smooth a sensor's (here: LilyPad Temp. Sensor) output

• The output values of sensors may vary according to the model no. The sensor’s datasheets can give information about the mapping of analog output values and sensed variables.

• For some use-cases you might want to cover the capacitive foil of a capsense setup with a thin insulating material (e.g. paper or plastic) to increase usability and to improve stability of the output values.
• You’ll need to reset the capsense variables at an interval of 100-300 loops to prevent it from accumulating tension and returning false positives. The code snippet below shows how to do it.
  
  [...]
  //Touch input
  CapSense cs_2 = CapSense(7,2); // 10M resistor between pins 7 & 2, pin 2 is sensor pin
  CapSense cs_3 = CapSense(7,3); // 10M resistor between pins 7 & 3, pin 3 is sensor pin
  int counter = 0;

  void setup()
  {
  // your set up
  }
  void loop ()
  {

  counter++;

  //CapSense Input

  long total2 = cs_2.capSense(30);
  long total3 = cs_3.capSense(30);
  currentMax = max (total2, total3)
  if (currentMax > 5000) {

  // do something, e.g. check which foil has been touched and trigger some activity

  }

  if (counter > 100) {

  cs_2 = CapSense(7,2);
  cs_3 = CapSense(7,3);
  counter = 0;

  }
}


[Capsense Sensing Library, http://www.arduino.cc/playground/Main/CapSense]

[Servo Library, http://www.arduino.cc/playground/ComponentLib/Servo]

Next steps?

The current prototype cannot regulate the air pressure due to the compressor’s two-staged control (ON/OFF)  and the construction’s inability to detain the insufflated air. As a consequence the hat’s insulating functionality only has two stages, whereas for a real-world application a stage-less regulation would be desirable. The prototype could be enhanced by the use of  seamless produced inflatable areas (e.g. made from silicone) in combination with two electromagnetic valves. This enhancement would allow to control the amount of air comprised in the hat continuously at any time according to the temperature sensor’s analog output or the user’s preferences.

A further drawback is the weight of the compressor. At the current stage the compressor has to be carried in a backpack. To make the prototype more user-friendly the compressor would have to be reduced to a minimum of weight. As previous tests proved, commercially available smaller devices, such as USB-hovers, do not have sufficient strength for this use-case. Therefore a device which combines both qualities – low weight and sufficient air pressure- would cause a significant improvement of the prototype’s usability and marketability.

Values and potentials?

In fact the idea to use air to insulate clothing is not new, duvet jackets use the air comprised in it’s feather filling to keep the owner warm. Nevertheless our idea is unique due to its ability to interact. In contrast to a duvet jacket, which is always warming, our headpiece is environment-aware and can be adjusted manually according to the ambient situation. By copying the bird’s anti-freezing technology we are able to create a cloth which is suitable for a wide range of situations. Our  concept can be extended to various kinds of clothing such as jackets or trousers. To determine the technology’s market value it would have to be proved if the puffing up of the cloth actually increases it’s ability to insulate. If that is the case we could think of possible application areas in polar research, astronomy or nautics.

A further aspect to be considered is the design factor. Though the headpiece contains a fully developed technologic system it still has to meet design aspects such as aesthetics and convenience. We tried to meet those requirements by wrapping the inflatable areas in a felt cover, which is cushy on the user’s skin and gives the headpiece a familiar hat-like look. Nevertheless it would be a challenge for designers to create new types of inflatable and attractive clothing.

AmbiLEON

Frederick Himperlich und Tanja Neumeyer

{lastname}@cip.ifi.lmu.de

What is it?

When you look at the chameleon in the picture, you see an animal climbing a tree having a green body and a red head.

If it’s trying to hide itself, it definitely failed!

During our research in bionics we learned about chameleons that they use their ability to change color not only for hiding but as a reaction to their surroundings. We were ecstatic about this fact and wanted to use this in our project.

Changing colors is the essential thing about our AmbiLEON-Prototype. Our idea was to develop a kind of lamp which creates ambient light in not static or random colors but in colors which express certain states or happenings.

How does it work?

When we began to build AmbiLEON, we knew the most important things were the RGB-LED-lights.

To create a random color with a RGB-LED you need 1 digital output for every color so three outputs for one LED-Strip. Since we wanted to use many LED lights, more than the ArduinoUNO could handle, we decided to use the Arduino Mega with a total of 14 PWM digital outputs Since the LED-Strips need a 12 V power supply we used transistors to control the 12V LEDs with the 5V digital output.

Now the base was given to let AmbiLEON change its color. The most important ability of our prototype was camouflage. To make that possible we needed the Ambileon to “see” its surroundings. For that we added a Camera to the Ambileon and wrote a little C# app that did the capturing. The captured camera pictures are separated in four columns (one column for each LED-strip), for each coloumn the average color is calculated and send back over serial port to the ArduinoMEGA which changes each LED-strip’s color accordingly.

Image 1: Camouflage mode

After developing the camouflage mode we went for the case. Our idea was to create something animal-like which reminds of a reptile. So we designed the “rips” for the insides in scale shape and four semicircular pieces which should serve as stage where we could stick the LED-strips to. We designed the main rips to hold the camera, we adopted the form of the camera to fit it exactly into the rips. All pieces look like an arc to provide a way for the cables inside the skeleton. To fix the rips and semicircles we also made a board where we simply fitted the rips into trenches of the same length.

Image 2: Designing the case

The material we used was acrylic glass cut by a laser cutter according to our design made with Adobe Illustrator.

Image 3: final case

Since there was some time left we decided to develop two additional modes for our AmbiLEON to show the versatility of chameleons. So the second ability AmbiLEON should have was a temperature mode. We added a temperature sensor which captures raw analog values. These values are mapped to the Celsius scale.***with the code, formula) AmbiLEON changes its color in steps of 5°C. Beginning with blue at temperatures < 0°C it can fade to different colors till it reaches red for temperatures beyond 25°C.

Image 4: copper foil touch sensor and camera embedded in case

The third, most interactive and, in our opinion, funniest ability of AmbiLEON is the pet mode. Chameleons show their emotions by changing colors and so shall AmbiLEON. It’s possible to pet AmbiLEON which reacts differently whether you pet it the right way or not. It is happy if you pet more rips at a time or single rips from the “head” to the “tail”. It shows its happiness by changing its color randomly between purple, cyan and blue in the area you pet and goes back to standard green if you stop. Additionally it vibrates to provide a haptic feedback and a sound which reminds of purring cats.

Image 5: vibration motor

If you pet AmbiLEON from “tail” to “head” however it will become angry which is expressed by a red color. Then it needs some time to calm down before it can be happy again.

This feature is realized by some copper foil stuck to the rips and connected to the ArduinoMEGA by a wire. Via CapSense it is possible to get higher values if someone touches the copper foil. To make it even more sensitive, like we needed to provide accuracy even with cloth between fingers and foil, we added a 10 MΩ resistor.

Now AmbiLEON was almost ready. The cables were fixed in the holes of our skeleton and we slipped some cloth (taken from some white tights) over the AmbiLEON skeleton to create the ambient light.

Image 6: AmbiLEON with skin

Values and Potential

During the course we already made the experience that people liked trying out AmbiLEON, petting it, seeing its reactions and playing around with the camouflage mode. The ability to play with AmbiLEON and the dynamic decorative element had an enormous attractiveness to people. We are extremely confident that AmbiLEON would enjoy great popularity due to the possible interactions, reactions and modes. In opposite to normal ambilight it doesn’t simply have a static sequence of colors but presents an exciting alternative to ordinary lamps.

We see AmbiLEON as a beginning of revolutionary furniture which responds in many ways to its surroundings and thus creates a whole new way of interior design possibilities.

Next Steps

The current version of AmbiLEON uses a button to switch between the three different modes. In future it shall be possible to union at least two of them, which would be camouflage and the pet mode.

Furthermore we would try to avoid the many cables that come out of the AmbiLEON by putting the ArduinoMEGA inside the skeleton together with the used bread board. Then the power supply and the camera cable would be the only wires to deal with. Of course in another step it should be possible to use the camouflage mode without a PC.

That’s how AmbiLEON came to life and we experienced the course: the three of us had a lot of fun.