Our project is an embedded system implementation that utilizes microcontrollers, servos, a camera, and a Nerf blaster. Put together, we have a Nerf Sentry that can serve as both a surveilance and entertainment device that fires upon detecting the presence of any person without the need of user input like a remote control.
Eduardo Rocha
Brandon Tran 
Sameh Fazli 
Luccas Chang Demo:
- Setup components in the similiar as the overall system diagram under diagrams. Specific wirings are annotated. For other fields, such as VCC or GND, you can wire them accordingly to your convienience it through a breadboard or through the available ports of an Arduino.
- Follow the steps in the readme of the Face_RecognitionV1 directory to set up the face recognition module and app notification side of the project.
- Using the Arduino IDE, upload the file named "Horizontal_Movement.ino" from the Horizontal_Movement directory into the Arduino Uno that is labeled "arduino uno r3 - 1" in the overall system diagram below.
- Using the Arduino IDE, upload the file named "Motor_Driver.ino" from the Motor_Driver directory into the Arduino Uno that is labeled "arduino uno r3 - 2" in the overall system diagram below.
- If they are not running, run the "client.py" and "server.py" files from the Face_RecognitionV1 directory. You sentry should now be able to detect any any one facing towards the usb camera.
The following two are state machines that we are able to implement to some extent
- Horizontal Movement:
- State Machine for Trigger Pulling:
- State Machine for Firing Signal:
Being an embedded system project, we do not have a graphical user interface or anything of that sort yet. In its place, we have a rendering of what we envisioned for our implementation at the very beginning of the first quarter.
In Milestone One, we have two major components in development: the Nerf blaster itself and the base/platform on which the blaster will be mounted on. Below is the dissassembled Nerf blaster with the components needed for it to power ON and fire. We also took away the need for the 6 C batteries needed to power the Blaster and instead attached a voltage booster to an AC to DC power supply. Decreasing the overall weight of our Blaster.
We begin addressing the need for horizontal movement in Milestone Two, approximately the latter half of the first quarter. What we ended up doing was cutting a wooden platform on which the blaster will be placed upon. We also cut another wooden board of similar dimensions and cut a hole in the center for a servo to fit through. In between these two wooden boards, we put a metal ring on both boards in which, in between, ball bearings are present. A wooden block is placed above the servo. This configuration assists the servo in moving the blaster above.
At around between Milestone Two and Milestone Three, the first half of the second quarter, we decided to build the main base which is what the Nerf Sentry will be planted on. The main base will look similar to an open box. The top of this main base is sandwiched with the base of the Nerf Blaster.
For Milestone Three, we place the components of the Nerf Sentry in a more organized fashion so that the wires are not all over the place and that most of our components are in one place. From then to Milestone Four, which is around the last half of the second quarter, deals more with being able to implement some form of Android notification and improving the performance of the face recognition, with one issue being improving the framerate. The images below show the project in around the second quarter with labels on some of the notable components, followed by screenshots of the Android app and the notifications it receives.
| 
Development
Install Node Package Manager (npm). Helpful Documentation
Dependencies regarding face recognition and the Android app notification are in the readme in the "Face_RecognitionV1" directory.
Pip Install opencv-python Documentation
arduino-libraries/servo Repository