Inspiration
We, as all students do, frequently and unwillingly fall to the powers of procrastination. This invention is for when the little cute Pomodoro and Screen Time reminders are a tad too easy to ignore.
What it does
The device sits in a predetermined area that you would not want to be in order to focus. For example beside your bed, on the couch, in front of your gaming PC/console. If it detects a person there, It will aim at you and fire projectiles.
How we built it
We built it by integrating a variety of technologies. Firstly, in terms of the frontend, it works with with an Android app developed using the Qualcomm HDK 8450 which has autonomous controls such as connecting to the projectile gun, turning on and off. The app also takes care of the ML Computer Vision needed in order to both detect people and where they are via Google's ML Kit. This then sends this information wirelessly via Bluetooth to an Arduino which is hooked up to two motors that control the aiming and firing of the projectile. The angle at which the projectile launcher turns is approximated with the user sitting 50-100cm away.
Challenges we ran into
We ran into multiple challenges during the project. Firstly, none of us had any experience developing an Android app and using an HDK8450, so we had a lot of ground to make up in order to start developing the app. Secondly, we found the Bluetooth module connection to be quite difficult to get working, as the official documentation seemed to be quite limited especially for beginners to Android development.
Accomplishments that we're proud of
One thing we are extremely proud of is the number of different systems and devices we got working together smoothly. From Computer Vision, to Bluetooth Protocols, to Arduino Programming and Mechanical Design, this project brought together a whole variety of fields, and we are proud to have been able to cover all of those bases as smoothly as we did.
What we learned
As beginners to Android development, we gained a plethora of knowledge on how to build, develop and deploy a working Android application. We also gained experience working with Arduinos, especially involving the communication aspects including sending and receiving information via Bluetooth. Finally we learned about deploying a working ML model in a solution of our own.
What's next for Failure Management 101
We would like to add movement by putting the whole mechanism on wheels to allow it a greater degree of freedom. We also had plans for voice control, as well as plans for the robot to have access to your laptop in order to determine whether the user is on non-productive websites. Finally, in a from a more realistic and practical purpose, we could envision robots like these helping in patrolling/guard duty as an aid to policemen, although perhaps not firing paper projectiles anymore.
Built With
- android-studio
- arduino
- bluetooth
- computer-vision
- hdk-8450
- machine-learning
Log in or sign up for Devpost to join the conversation.