The Fruit is a robot that sets out to solve the problem of finding adequate housing and lighting for indoor house plants. In many houses plants do not have enough light, and they do not get even light on all sides, so their growth is stunted and uneven. The Fruit solves this problem by traveling over a given course of acceptable “parking areas” to measure for light brightness, and then parking in the area with the brightest light. It then slowly rotates for the duration of the plant’s life span to ensure even growth on all sides.
We wanted to create The Fruit to optimize plant growth with personal robots, since this is an emerging technology with increased accessibility. It is becoming cheaper and easier than ever before to access and build these robots, so they seemed like the perfect way to address this common household problem.
The Fruit got its name from its two main components: the rind and the core. The rind is the superficial outer layer. It houses the plant and covers the electronics, and is mounted on a gear that turns continuously at all times to provide even light for the plant. It has two sub-components: the pot, and the windows. The pot is mounted on the rind and contains the plant, while the windows are small acrylic areas embedded in the top of the rind that let light through to the phototransistors for measurement.
The inner layer is called the core. The core contains motors, electronics, and phototransistors. This layer has a base with laser cut holes to house the arduino, phototransistors, and wheels so they fit perfectly. Above them is the battery pack, motors, and gears. There are two arrays of phototransistors: one facing downwards to track the line so the robot can stay on course, and one facing upwards to sense lighting conditions.
We started out with the idea to have the robot not have a front, but rather a circular array of phototransistors so that it could align itself irrespective of the orientation. This ended up needing far more phototransistors than we would otherwise need, so we scrapped this idea for the next iteration. The next iteration took a more traditional approach of having a front in order to only need 5 or fewer phototransistors, but then we needed to figure out a way to make the plant spin. This is when the idea of the rind and the core came into play. The first iteration of the rind was made up of many stacking wood pieces which made it very unstable, so we switched to flexible wood. This made it more stable and more aesthetically appealing. Once this problem was solved we moved onto decisions about materials, deciding what to make the planting pot out of. The first iteration made out of plastic was not very smooth, and had sharp edges, so the next iteration was made out of 3D printed wood filament that we could sand.
When we set to focus on the electronics we found that it took more effort than we anticipated, so we ended up using only one phototransistor on the bottom instead of an array, and put the phototransistors on the top on hold for now. The current setup works for line tracking, but is slightly jerkier than it would be if we had used in array. In future iterations we would use more phototransistors and finish the top part to take in the lighting of the room.