Social Creature Interactive Sculpture
(Work in Progress)
Self Directed Project Created at Laney College
This art piece aims to mimic life through a dynamic system that responds to its social environment. As social creatures, we are constantly effected by those around us. The physical reaction I experience throughout my body during both times of connection and times of anxiety and panic inspired this goal. In this project, I attempt to take the essence of this physical sensation and bring it into an abstract form. This form consists of a porous spherical outer shell, a soft internal body, and a mechatronic system that brings the structure to life.
Mechanical System
Current State
The mechanics primarily consist of a large, rigid, linear actuator. This contains a shaft that connects to a motor which drives a lead screw connected by a belt drive. This takes the rotational motion of the motor and converts into linear motion of a flat plate while giving it mechanical advantage (compared to a linkage system).
In the above image, the plate that interacts with the soft body is shown at the bottom. The motor connects to the shaft on the far left. The two shafts that run through white bearing guides act as stabilizers to keep the system square and rigid. The remaining shaft is a threaded rod that is forced forward and backward (up and down in this image) by the matching nut that sits inside a sub-assembly. This sub-assembly connects to the nut, bearings, and belt, shown in the image below. The bearings, shafts, couplers, and lead screw were purchased. The mounting plate, belt sprockets, lead screw sub-assembly, linear bearing modules, and motor mount were designed and fabricated by me using digital fabrication tools including a 3D printer, laser cutter, and water jet.
This image shows the lead screw sub-assembly in detail. These two (white and black) 3D printed components sandwich the brass nut of the lead screw, directly connecting it to the 3d printed sprocket that is driven by the belt. These components also contain a ridge that centers it between two bearings when assembled in the full system.
Process
The process for developing the mechanical system started with material experimentation. Using a variety of soft body materials, I pushed into them with different objects, seeing how that would effect the overall shape. I found that a flat plate pushing against a yoga ball created the response that I was looking for, so I began designing around that. I initially started to build a rough prototype to give myself a sense of the dynamics I was looking for.
After experimenting with this idea, I drew an outline for the mechanical system that would use various components to achieve the motion I wanted.
This outline became the starting point, allowing me to organize and order all of the components I needed. With the components in hand, I was able to design, print, and cut all of the connecting pieces. This led to my first design iteration:
This assembly worked as an initial proof of concept for the overall design. The plate moved as expected, the tension in the belt worked, and the pieces all fit together as expected. While this was a solid starting point, there were many issues that immediately surfaced. The belt rubbed against the base plate, the nut on the lead screw wobbled and created unnecessary friction, and the linear bearings I had purchased were low quality and falling apart. This led to the improved lead screw sub-assembly (shown earlier) that used 2 bearings to reduce wobble, and the replacement of the linear bearings with my custom designed linear bearings. In the next (and current) iteration I also reduced the plate size and raised the input shaft to allow for the motor to fit.
Electronic System
Current State
The electronics are controlled by a Raspberry Pi (RPi) computer to allow for image processing and audio output. As the RPi is not running a real-time operating system, an Arduino microcontroller is used to control the steps for the stepper motor. The microcontroller talks to the RPi via serial communication, and the microcontroller drives the stepper motor using a current-limiting stepper driver. The microcontroller also has input from two limit switches that allow the system to home itself and prevent damage. The RPi takes images using a small USB camera, and outputs audio to a headphone jack. The whole system is powered by a single power supply that connects to an outlet through a power switch and IEC connector. The system is cooled using heat sinks and a central fan.
The above image showcases the current state of the electronics wiring. While there are many wires centered around the stepper driver, they are packed into separate bundles, and are connected using JST connectors when possible. Now, the electronics are set up to allow for testing and iteration of the software, but later they will be installed and organized on a mounting board that will fit inside the project's enclosure. The cooling system has been purchased, but not yet assembled.
Process
As I have more experience in mechanical engineering than electronics, this system included some new areas of learning for me. While I knew how stepper motors worked and were operated, this was the first project of mine that included them. This hands-on experience introduced me to many quirks and considerations when designing a system with a stepper. The main insight came from purchasing the wrong stepper driver. My original iteration of this system (shown below), included a simple dual H-bridge motor driver. While this theoretically works and allows for the correct voltages to be applied to the motor, it does not do anything to account for the electro-magnetic force created by the motor spinning (back EMF). This means the motor is getting inadequate current to drive effectively which causes random stopping and reduced torque.
In addition, with a current limiting driver, the motor can take far higher input voltage, allowing it to operate effectively at higher speeds. The power supply in the first iteration powered 2 amps at 12 volts. The current supply gives 10 amps at 36 volts. While this is certainly overkill, it allows me to ensure that all of the components will have the power they need, and the motor can run as effectively as possible.
Software System
Current State
The software for this project is currently in an earlier phase of development. So far, I have written software to test each of the components separately, but have yet to link the pieces together. I first created software that used an acceleration curve to drive the stepper motor in a sinusoidal motion (to mimic breathing). I then wrote a script that used a computer vision library to take a photo with the USB camera and analyze it to count the number of faces seen in the shot. This worked pretty reliably in a well-lit area. I then wrote a script that uses the limit switches to home the motors. This worked well, as it responded to the limit switches seemingly instantaneously. Lastly, I wrote a script that used an audio library to output sound from stored files. Next, I will setup communication between the RPi and the microcontroller. Finally, I will interlace all of these scripts to run in sync together.
Ceramics
Current State
The ceramic piece for this project is the visual center. It brings structure and form to the entire sculpture, and it allows the soft body to create interesting forms in their intersection. The piece is currently in the process of its second firing. All of the clay and glaze work is finished, and I am awaiting the final result. Below displays the most recent state of the piece both before the first firing (left), and before the second (right).
The form is meant to remind the viewer of organic, living beings while remaining separate from any real organism. The intersecting coils were inspired by both animal ribcages and tree roots. The texture mimics woody branches, while the overall form feels alien.
Process
This piece began with some iteration and testing. The first conceptual design was created as a test to see how the soft body would interact with intersecting coils. This is shown below.
This form succeeded in making interesting interactions with the soft body that felt organic and connected. However, the texture and pattern of the coils felt somewhat void of life. In this new version, I solved those issues by creating groves, patterns, and textures that flowed through the coils. I also decided to use a more earthy clay to make it feel more natural, and to stain it with oxides to emphasize the texture. I also opted for a spherical design to allow for the whole structure to interact with the soft body together. To make it feel like a present being, I also made it far larger (roughly 14 inches in diameter).
My next challenge was to set a plan to construct this complex form. I decided to use a bowl to support the project as I built it inverted. This worked very well and kept things together nicely.
I then encountered several issues with cracks forming, and some coils breaking off during the drying stage. While disheartening, it forced me to learn some new tricks to repair cracked clay. After the joints were successfully mended and the clay was leather hard, I was able to take it out of the bowl and create the texture.
The texture work involved using sculpting tools to carve grooves into the clay. While doing this, I visualized each coil as a branch, and allowed the shape of the form to guide where these branches separated and combined. These connection points allowed me to flow the texture between the branches, connecting them to each other. Deep groves were also added to split the branches and add a vein to follow across multiple different coils.
Next Steps
While this project is pretty far on its way, there is much more work to be done in the near future. The largest area of exploration will come in the development of the soft body. Currently, I plan to use a small yoga ball that I under inflate to allow for much change in shape as the plate pushes and pulls it. I then plan to cover that ball in a lining sewn together from a variety of fabrics. I still want to explore this further though, as I want to ensure that it keeps the organic elements in tact. Having a fleshy silicone body instead may appear more organic, but could also lead to more feelings of fear that go against the goals of this project.
In addition, the enclosure and mounting system for this project is yet to be designed. I currently plan for the project to be mounted to a wall, protruding out from it, but that concept may change after putting the system together. I plan for the enclosure to be minimal and hidden, to allow for the organic form to take center-stage.
The sound design for this project has been started, and content has been created and organized, but there is still much work to do on that front as well. I want to blend the peaceful feeling of connection into the anxious panic of overstimulation seamlessly, which will require more testing and experimentation.
Overall, I am happy with where the project is at, and incredibly excited for where its heading towards.