IloMilo/Spring/2020
Tripping on Shrooms – Outer Shell Design
Author: Farland Nguyen
Table of Contents
Introduction
Aesthetics provide some semblance of order to engineering solutions. Creating a clean and neat appearance to any design not only appeals to the eye but it can also force systems to be designed in a more user-friendly manner. Instead of a mess of wires, a product can be designed with wire management in mind to create tidy wire routing that clearly shows the systems’ connections between each other. With IloMilo, our design aesthetics were more focused upon looking cute, with system management as a secondary desire. IloMilo’s design required the concealment of two major systems: a Paperbot and an internal joystick that both provides movement to the overall product. The final outer shell design agreed upon required a starting design point that morphed into a basic silhouette to achieve this task.
Germination: Alonso Martinez’s Mira
Figure 1. Alonso Martinez’s Animated Robot
As the main inspiration for our internal systems, it would be thoughtless as to not attempt to use Alonso Martinez’s outer shell for his Mira robot. As seen in Figure 1, Martinez’s Mira shell consists of two half-spheres connected by a joystick, with the bottom sphere being flat. On the outset of our project, this outer shell design appeared to have everything going for it. Mira’s shell design was mature, obviously created through several iterations by Martinez, and had the space required to hold not only a joystick, but the OLEDs and ToF sensor within the head. Mira’s outer shell was also compact enough, seen in the vidoe, to house all of Martinez’s systems. The closeness of the two half-sphere would have definitively decreased the length of wires required in routing the OLEDs and ToF to the 3Dot board of the Paperbot we agreed as our drivetrain.
Be that as it may, this is where Martinez’s Mira outer shell design began to show its limitations for our purposes. While efficient, the Mira shell was created around Martinez’s systems and not around the systems we planned to incorporate. Integrating the Paperbot into the Mira shell demanded an increase in the size of the lower half-sphere, in turn, required a larger upper half-sphere. At this point, the servos required to move the joystick had been decided, adding in more space that needed to be accounted for. With the Paperbot and the servos, the Mira shell would have to be, at the very least, three times the size of the original Mira. At this stage, due to the height of the Paperbot, a massive Mira shell would have wasted a tremendous amount of space around the Paperbot. The 3Dot board is located inside the Paperbot in such a way that connections can only be made through the top of the Paperbot. Any attempts to rectify this issue and maximize the volume efficiently of the Mira shell called for extensive redesigns of the lower half-sphere or Paperbot or both which in turn required weeks, if not months, of CAD work. In the end, the Mira shell design did not offer what we needed in an outer shell within the time frame we had. .
I had looked at imitating the characters from the game IloMilo, but was discouraged by their design. They were gumdrop shapes with front-facing area that represented their faces. This would require changing the joystick from a vertical arrangement to a horizontal one, as there were no other areas to create head movement, adding a level of complexity that provided no benefit. The faces moving would have also left holes where the internals could be more easily viewed. Thus, I turned to a custom design of my own to achieve our tasks.
Fruitbody Selection: Final Mushroom Torso
The largest issue the Mira shell faced was maximizing the space taken up by the Paperbot. More specifically, the empty space created due to the height of the Paperbot inside of a half-sphere. To eliminate this, I decided to eliminate the half-sphere torso and replace it with a cylindrical torso. A cylinder created a more vertical appearance that would allow the tall Paperbot to seamlessly meld into the torso. The curvature of a cylinder would also provide additional space if required. If a rectangular prism torso was used, the Paperbot could have had a tighter fit with the torso. However, when fitting a rectangular object in the round ends of a cylinder, the cylinder must be slightly larger than the object to achieve this. Therefore, even with desire of a small IloMilo, there is still potential space for wire management or additional systems. The joystick and the servos associated with it would also have much more room to work with, freeing up some design constraints in this area.
At this point, I recognized that this torso had taken the shape of a stem, seen in Figure 2. With the edges rounded off and the addition of a slight taper to the cylinder, this new torso created a stubby base that could appear cute but not become the focal point of the bot. The stem torso was also ambiguous enough to have any sort of head attached to it. A plant or animal head would have easily fit over the custom torso.
Releasing Spores: Final Mushroom Head
With an ambiguous torso, I turned to designing the head. The head could arguably be considered the most important part of the custom outer shell. It would provide the movement to express the mood of the IloMilo bot. Not only that, the head would also have to be sturdy enough house two OLEDs and a ToF sensor as well as be light enough to be supported by the joystick. In the interest of lowering complexity, I turned back to the Mira shell design. By using a half-sphere, seen in Figure 4, and modifying it to be slightly wider at the base, I could create a mushroom head to be placed on top of the stubby stem.
The mushroom head was chosen mainly due to the simplicity of the solution. It could be achieved by stretching the base of the half-sphere and adding two eye holes. External physical modifications with props, such as headbands, bow ties, or cat ears, added an extra level of customization to differentiate the bots from each other. The mushroom head creates a basic silhouette that can be altered either permanently or temporarily, simplifying any potential future designs. The mushroom head shape would also provide an enormous amount of space for the placement of the OLEDs and the ToF sensor, slightly subverting the desire for a small bot.
Conclusion
When assembled in Solidworks, the finalized outer shell is able to house the custom joystick without revealing its existence, seen in Figure 5. Once printed out, the Paperbot was able to fit within the recess at the bottom of the torso. The interior of the head was not only large enough to house the ToF and possibly the OLEDs, it had extra space to provide for extended cables for both systems to compensate for the degree of movement from the joystick. All in all, the final design was able to meet most if not all of the requirements created for it.
However, there are a few characteristics that could be improved upon. To fit the joystick within the stem, the joystick had to be quite tall. Adding onto this, the head is around 400 grams, causing the center of gravity for the outer shell to be taller than expected. Since the outer shell is top heavy and is placed on top of the Paperbot, the entire bot has a higher chance of being tipped over. One potential solution would be to replace the Paperbot with a dedicated, fully-integrated drive train into the torso. This custom drive train, if designed correctly, could shorten the torso and bring the center of gravity down.
References/Resources
- https://youtu.be/0vfuOW1tsX0