Cycle

• 

  Cover

Inspiration

What if we could create a connection between manufacturers and consumers by providing information regarding product recycling to the consumer and connect them with available recycle deals? The vision we have is that the user will be able to easily capture an image of any object in their possession using AR glasses and receive information on how to recycle, where to recycle, as well as potential trade in programs, reducing the friction between intent and action.

How it works

Our app utilizes the Snapdragon Spaces SDK for Unity on the Lenovo ThinkReality A3 smart glasses connected to an android phone. Using the SDK, we allow the user to capture an image using the glasses. The image is sent via HTTP to a python-based web server that queries an unofficial Google Bard AI API with a set prompt to generate information regarding how to repair or recycle the item. The information is then displayed in the AR space, while the user experience is controlled using a smartphone touch screen.

How it was designed

Concept and design: We focused our efforts on the UN Sustainable Development goal of responsible consumption and production. Achieving economic growth and sustainable development requires that we urgently reduce our ecological footprint by changing the way we produce and consume goods and resources. We decided we wanted to address the end of the life of an object before it goes into the waste stream.

According to the US EPA, "the total generation of municipal solid waste (MSW) in 2018 was 292.4 million tons (U.S. short tons, unless specified) or 4.9 pounds per person per day. Of the MSW generated, [only] approximately 69 million tons were recycled and 25 million tons were composted" (https://www.epa.gov/facts-and-figures-about-materials-waste-and-recycling/national-overview-facts-and-figures-materials).

"According to the Institute of Scrap Recycling Industries (ISRI), 66% of Americans would not bother to recycle a waste item if it isn’t easy to do so" and most people fail to follow through with the desire to recycle or recycle properly due to a lack of knowledge regarding what items can be recycled and the state they need to be in in order to be recycled (i.e. clean, not contaminated with food) (https://cleanrobotics.com/are-humans-the-biggest-barrier-to-recycling-how-does-ai-help/).

In order to reduce the amount of waste going into landfills, we decided to focus our efforts on providing easily accessible information regarding recycling in an effort to reduce the friction between having the initial desire to recycle a product, and the actual action of doing so.

Challenges we ran into

Technical Development: Because Google Bard does not have an official API or a Unity plugin, we had to create a web server using python code from an open source python package that queries the google bard API. This combination of having to rely on writing Unity code to communicate with an HTTP web server as well as having that web server use an unofficial API created a multitude of issues that dropped the success rate of image analysis. While writing Unity and python code that could get a Google Bard response proved to be time consuming, once we were able to more or less consistently get responses, the challenges weren't over. We had to engineer prompts that could provide more consistently formatted answers, and write parsing code that could handle the variation of response and display info in an organized way in AR.

User Experience: Since team's designers had no previous experience with XR, visualizing and writing out the logic for the flow of the application and thinking how to make the navigation as intuitive as possible was challenging.

Accomplishments that we're proud of

First, we are proud of the journey we took as a team. From being total strangers to each other, we became great friends that were able to create and execute the project we envisioned. Second, we are proud that we were able to make our demo application work and figure out how to connect the AR glasses to Google Bard without an official API. We are also proud of the flow and UI our designers came up with.

What's next for Cycle

There are several important features we want to implement in the future:

Visualize the outcome: By scanning an object, users can see the real-world environmental impact of recycling it. For instance, scanning a plastic bag might show how much lumber could be made from recycled bags, providing a tangible sense of the positive change.

Repair and revive: The app's AR deconstruction and play features could inspire users to give new life to old items. Imagine virtually disassembling a chair and customizing it with new parts, or using AR to turn a broken vase into a unique planter. This interactive experience could foster creativity and appreciation for repairable items.

Overall, Cylcle's potential lies in using AR to connect users with the consequences of their choices, motivating them to make more sustainable and resourceful decisions. By seeing the immediate and tangible impact of their actions, users are more likely to embrace recycling, reuse, and repair as ways to contribute to a greener future.

Built With

• adobe-illustrator
• figma
• qualcomm
• unity