From our concept tests and early interviews, we learned that our users do want to correctly dispose of things but lack the knowledge and means to do so. That's why we provide a better and literal transparency that not many other smart recycling bins or recycling centers provide. Our bin has transparent windows allowing users to see RightCycle actively sorting of their waste. Through our optional app, users have even more opportunity to see where their waste ends up. These levels of transparency and the new technology available are things we hope to take advantage of and will be discussed further in the feasibility. Overall, our goal is to provide convenience and guarantee to our users.
We began with using material sensors to detect the waste for sorting and using bluetooth connection to the users’ app to optimize convenience and speed in the process of using our product. After presenting our initial conceptual design to the class, we got some feedback how bluetooth auto-connect raises privacy concerns, and through our own research, came to the conclusion that material sensors cannot detect a wide variety of waste items and are expensive.
After looking into camera detection paired with AI learning as well as NFC scanning (similar to Apple Pay), we opted to implement these changes in our prototype. NFC scanning protects the privacy of the user and doesn’t automatically connect to our server without the user’s consent. The AI technology with our camera is much more efficient as we can access multiple databases of common waste items and constantly update the sorting mechanism. Using regular material sensors restricts us to certain types of materials and thus limits our accuracy level in sorting waste. To optimize speed and convenience in using our product, we noticed most of our competitors only allowed a couple items to be thrown at a time. With a conveyor belt system, users can dump all their waste at once but the disposal would still be staggered. Adding 1 laser sensor to detect each internal bin’s capacity, we also optimized pricing and effectiveness for our customers.
For our app design, we got some feedback from our test survey to make our users more involved with the act of proper recycling and that monetization wasn’t a common need. Thus, we removed the monetization aspect and pivoted to a point system using an interactive game, RightCycle visits, and correctly sorting outside of our product to earn rewards.
After looking into camera detection paired with AI learning as well as NFC scanning (similar to Apple Pay), we opted to implement these changes in our prototype. NFC scanning protects the privacy of the user and doesn’t automatically connect to our server without the user’s consent. The AI technology with our camera is much more efficient as we can access multiple databases of common waste items and constantly update the sorting mechanism. Using regular material sensors restricts us to certain types of materials and thus limits our accuracy level in sorting waste. To optimize speed and convenience in using our product, we noticed most of our competitors only allowed a couple items to be thrown at a time. With a conveyor belt system, users can dump all their waste at once but the disposal would still be staggered. Adding 1 laser sensor to detect each internal bin’s capacity, we also optimized pricing and effectiveness for our customers.
For our app design, we got some feedback from our test survey to make our users more involved with the act of proper recycling and that monetization wasn’t a common need. Thus, we removed the monetization aspect and pivoted to a point system using an interactive game, RightCycle visits, and correctly sorting outside of our product to earn rewards.
Initial Prototyping:
Before creating our 3D CAD model of RightCycle, we used cardboard boxes to represent its overall size and the main physical components (opening and division of sections).
Test & Learning Cards