Mechanically speaking, the design of the sorter is relatively simple. Most of the movement in the sorter will be done through motors and involve rotations of conveyors. As seen above, the group did perform a technical analysis on the inclined stabilizer to discover the maximum angle of incline allowed for the LEGO pieces to not slide down. The prototype then confirmed these calculations by performing at an angle of approximately fifteen degrees. Future designs and prototypes will reflect these results by lowering the inclined plane to the fifteen degrees implemented in the alpha prototype. Additionally, the length of the inclined conveyor will be increased to ensure the LEGO still have enough vertical clearance to pass through the rest of the system.
The other minor mechanical aspect of the design is the vibration plate. The vibration itself can be achieved with vibration motors and suspension can be designed to prevent movement throughout the entire system. The most difficult mechanical part of the vibration plate will likely be its assembly.
The final and most complex mechanical aspect of the device will be the actuators. While the actuators will be activated by the code, they still need to move through mechanical means. Currently, the team plans to incorporate a small cam design that will slowly rotate and cause the actuators to open up. The group will spend more time to test and design the movement of the actuators in the future phases.
Due to the relative simplicity of the mechanical aspects of the design, the sorter will not require as much cost for production or as much effort in manufacturing.
The electrical design for the team’s final iteration of the product puts an emphasis on low cost, readily available components to reduce any extraneous effort spent on designing novel parts and reduce overhead cost with producing the device. This is primarily achieved by purchasing and utilizing off-the-shelf parts. The DC motors, vibration motors, and LED lights are all widely available and follow stringent government regulations and standards that ensure safety, uniformity, and effectiveness. Furthermore, this allows the team’s design to not require any special components; this increases the reproducibility and modular nature of the electrical parts.
Two major components that are not covered under the low-cost parts mentioned above is the Raspberry Pi V2 Camera and the Raspberry Pi Computer. These two components comprise more than half the estimated cost of producing the team's device. Raspberry Pi V2 Camera is a device highly optimized for image recognition and machine learning applications for developers. Raspberry Pi computer allows the team to code that will be compatible with the image recognition of the Raspberry Pi V2 camera and function as the software backbone of the device. Both of these components are fabricated by highly respected and highly specialized companies. This gives the team the confidence to utilize these sub-components without the speculation of failure, ineffectiveness, or incompatibility with widely accepted standards.
The software design of this device is integral to the functional capability of our team's aspirations. The team's design of the software puts an emphasis on convenience for consumers as well as using image recognition tools and software platforms the team is familiar with. This is achieved by using a Raspberry Pi 4, a small single-board computer, along with a Raspberry Pi V2 camera that is compatible with Tensor Flow, an open-source machine learning image recognition platform. A logic framework is shown above that highlights the higher level software flow as well as the interactions with the mechanical subcomponents. A preliminary software analysis is displayed above that demonstrates the potential to distinguish unique objects in the cameras image capture.
This software will need to be optimized to recognize the six distinct, archetypal LEGO bricks. The team will achieve this goal by "training" the logic to recognize features of the aforementioned bricks on a white back-drop "Image Recognition Tunnel" that will be flooded with light. An operator will be required initially to determine whether the features are shared amongst a certain piece type. Once the software has "learned" with a high certainty and accuracy, the system will autonomously recognize and filter to each brick's respective collection bin.
The team has designed this product with the primary purpose of simplifying and streamlining processes the customer segments already conduct. This, however, is not the sole objective, as the team is aware of the possible unintended consequences of using materials or processes that may be harmful or damaging to the environment. Being cognizant of environmental concerns, the prototype models are currently being constructed using balsa wood and cardboard which are very environmentally friendly material choices. The team intends to continue using these material choices in the ongoing prototyping of sub-components. In the scenario of mass production of this product, it is probably financially irresponsible to continue using the prototype materials. The team would then be required to conduct an environmental analysis into ABS, PLA, or PC plastics and associated manufacturing processes such as injection molding. Manufacturing on this scale using plastics notoriously harmful to the environment poses a significant challenge.
An important design factor the team has taken into consideration is the accessibility of the product. Keeping in line with expectations and quality standards of LEGO products, the team aims to create a product that is intuitive, easily setup, and does not require in depth technical know-how. This ensures the product is viable for all ages and all academic levels. The team’s dedication to the customer experience warrants us to create a product that is durable and long-lasting. The device should be able to withstand thousands of cycles without failure to maximize the utility created. These design considerations are taken to positively affect all groups interacting with the team's product: from the consumer to the manufacturer and the larger society.
During any engineering design process, it is the responsibility of the team to consider different ethical implications that may arise from a product that is a consumer-facing material. For the team, this is especially important because the LEGO sorting device will be established firstly, in classrooms, with professors, and in the collegiate community. The device is designed using many different moving mechanical components and these may pose a hazard to the operator of the device; most likely a student, teacher’s assistant, or professor. With this in mind, the team proposes a device that minimizes operator interaction and prioritizes autonomous functionality for the device. Furthermore, the moving parts of the device should be covered in order to minimize the risk of injury for anyone using the machine. Another area of ethical concern is deep learning technology with the team's deep learning camera approach to image processing. With privacy issues becoming a ubiquitous concern with technology and the product relying on deep learning methods, it is important to the team that there is security for the software and rapport with the customer segments to ensure a digitally safe and secure device.
One of the best aspects of working with LEGO is the product safety that The LEGO Group corporation is able to offer to its customers. The team is designing a product that works intimately with LEGO products and as such the standard of safety for the product should aspire to be as comprehensive as The LEGO Group’s. The various testing of chemical, physical, electrical, hygiene, and flammable properties that LEGO products undergo is state of the art. Also, similar to LEGO products, the team’s design for a LEGO sorting device should be accessible to all ages.