Abstract

The project explores the world of the Automata, in particular 18th Century precedents and vivisections of found toys. Through the reverse engineering process, a focus in research emerged: the uncanny. My initial focus was on the movement of these Automata, and what was it about the actions of these somewhat aesthetically life like figures that evoke such a ‘pleasurable’ response from the public. For me it’s this uncanny movement of the automaton that responds to our psyche. Freud and Jentsch define the uncanny as a response to the unfamiliar, the ‘unhomely’, a fear of the unknown. Freud relates this human response to a human repression of familiar elements of our childhood that are now reintroduced out of context. A human response is seen in public reaction to Descartes’ mechanized doll of his late daughter, and the talking doll of Thomas Edison. The following project explores these issues of uncanny and artificial intelligence through the production of a ‘monster’.

The monster is a working model for acquiring random, self-motivated movement. Beginning with a collection of toys, the parts have been reduced to the simple gear and motor cases that will create their movement. Photovoltaic sensors are used to stimulate movement within each device, from its exterior environment. Interfacing becomes a lesson in switches, where my project focuses on the production of tilt switches to create a variety in movement. The switch becomes a synaptic charge within the beast, in the hopes of stimulating self-directed movement. Further development focuses on the creation of a proscenium for the device, too work towards allowing the ‘autistic child’ to interact more deeply with the inhabitants of its space. My interest lies in the uncanny response, and how this issue of interface can be applied to architectural form. I have chosen trace experiments and analysis as one way of exploring the relationship between beast, its stage and the site around it.

Review

I received some really good feedback from the review. First was the discussion of how the device appears uncanny. There were suggestions that i should be using thermal sensors, instead of light sensors. The thermal sensor makes more sense for human detection. Ken Gregory recommended full use of the Arduino and to create pauses within the device to give the appearances of thought. I am not sure if i will abandon the switches just yet, but the pause i believe are crucial to the uncanny nature. The switches i believe are capable of pauses much more realistically than an Arduino. The randomness of the device has been a bit lost as i have pushed it further along, apparent in the proscenium. My interest lies in the uncanny, and its discussion about interface within architecture. I am currently reading The Architectural Uncanny: exploring the psychological reaction and relationship between occupant and space. Perhaps i can look into the relationship between the subconscious and the space we occupy, exploring in more depth about this duality, Cartesian dualism, and conflict between what Freud refers to as our repressed thoughts and our current situation.

More analysis of the movement, and the space it creates are my focus now: more importantly the relationship between the outside stimulus and the device tracing . The layer of interface between these two is what i want to understand more to develop how the structure can influence us.

Tracing movement

When i started to trace the movement, some interesting things came up. When the device moves its creates a clean geometric pattern, sometimes broken up because of the stopping of the arms ( due to too much weight on the arms) . When the proscenium is moving, the traces become random and chaotic. Some of the photos turned out nice when the proscenium begins to move (not just along the linear direction of the suspension line) along the upper suspension lines. The patterning is still circular through space but not symmetrical or predicatable, which is interesting. Also the tension movement, the bounce due to the object's weight, shows shifts in the geometric patterning: revealing a response of the device's movement to the prosceniums tension.Further exploration needs to look into the human interaction with the trace movement.

Proscenium up and running

The proscenium is now operational, but still a few glitches. Weight causes friction between the wood axles at each gear. When the pole is at 45 degree angle to the double suspension line above, friction between the bracket and pole slows and halts movement. Lubrication was a key to resolving some of those issues. The power, amperage, another issue, was resolved by wiring a separate power supply to each motor. In the future a more efficient gear system will probably help. I started to get the Arduino programmed and trying to figure out a way to reverse polarity on the motors through Arduino. I looked at using relays, using the Arduino to switch the polarity through a blink program. My decision to go this route instead of the switches is due to the nature of the tilt. I guess i could just abandon the floor device all together and use its inputs for the upper motors. I also used the extra tilt switch i had from before, located on the tail, to interrupt the device pulley along the suspension line, creating some pause input into the devices movement. Not only is the random important but the pause as well, something I noticed from earlier: I have been trying to replicate with switches all this time. The light sensors provide some interruption in movement, some pause, but its so direct that the uncanny gets lost.

Simplifying the proscenium

After puting up two suspension lines to carry the main suspension line rod, i started to notice the littlest vibration carry through out the system. What was interesting was the movement of the vibration, you could see it carry back and forth from one suspension line to another through the movement of the suspension rod. Perhaps the feed back of the structure might not be from the device but surrounding movement in the building.

Here are some pics showing the device that will carry the suspension line and device. One thing i noticed was that when i used a battery vs the power source at 9 volts, the power source had better success. Perhaps half the problem with the lack of movement is not enough amperage. Now I just need to reconfigure the switching so there is more movement along the frame.

Building the suspension cable frame at first i was concerned with the bow on the rod after adding tension to the cable. I went and got a thicker rod, but still some bow. But i think no matter what i do, the tension will take over, so instead of fighting it, I decided to embrace it. The vibration on the tension cables may interupt the device's patterns.

The up and down movement is still to be worked out. The wench idea was thought of, but the motors are not powerful enough. My old idea of the scissor lift would work but it was quite cumbersome and bulky. I think focusing on what i have now is more important, the up and down can be added perhaps in a couple days to add another dimension to the patterning,

Simplifying the proscenium

After cutting pieces for the upper mechanics of the proscenium i discovered some complications. It is a bit chunky and complex in some areas, so i decided to simplify. The suspension cable will consist of a rod with three cables in tension, and the pulley device will be repeated twice above to move the suspension frame side to side along perpendicular suspended cables. The up and down movement will be localized at the pulley device, below it in a wench like form. The overall result should be less weight and more refinement in the pieces.