🚧 Project EPES 🚧

An open-source water dispenser using Raspberry Pi / Orange Pi utilizing Edge-AI technology for resource-constrained devices. (Code, Circuit Design and Casing Design will be remodeled soon, project deadline on July 2025, Personal Gathered and Curated Image Dataset will be provided also). See early prototypes here (Still gathering images with Roboflow and using proper curation with FastDup).

Target Features:

• Automatic Dispensing - Automatically dispense upon container recognition as well as stop operation upon closely full of the said container.
• Compatibility with Orange Pi One
• Customizable and Hackable like Mugsy

Installation

Installing Libraries (for both OS)

• ESP32/8266 Board
• Libraries:
  • OneWire
  • DallasTemperature
  • HX711
  • TFT_eSPI
  • GUIslice

Required Software

• Arduino IDE
• GUIslice Builder
• Image2c:
• Blender (Optional in Windows)
• Fritzing (Optional in Windows)

---

Raspberry Pi Setup

Speaker On Boot (Needs confirmation)

1. List available sound devices:

   aplay -l

   (Identify the card number corresponding to the headphones for the speaker.)
2. Edit the ALSA configuration file:

   nano ~/.asoundrc

3. Add the following configuration (adjust the card number if necessary):

   pcm.!default {
       type hw
       card 2
   }
   
   ctl.!default {
       type hw
       card 2
   }

Script on Boot (Needs confirmation)

• (Instructions for setting up a script to run on boot can be added here.)

Rebooting the Raspberry Pi

sudo shutdown -r now

Note: It is recommended to use Remote Desktop to the Raspberry Pi for easier debugging.

---

Windows Setup

• Software Installation: Install the required software listed above.
• Library Installation: Ensure all required libraries are installed.

---

Setup

1. Clone the Repository:

   git clone <repository-url>

2. Navigate to the Project Folder:

   cd <project-repo-folder>

3. Prerequisites:
   • Main Materials: ESP32 and Raspberry Pi 4B (also applicable to Orange Pi One)
   • Circuit Design: The design is provided in Fritzing format in Circuit Design/Circuit_Design_EPES_v2.0.0.fzz.
   • Libraries: Ensure all necessary libraries (listed above) are installed.
   • Python Prerequisites: Located in Code/Raspiberry_Pi/src (refer to the requirements.txt file for installation).
   • Casing: The 3D model is available in the 3D Model/Project_EPES_v1.0.0.blend (Blender file format).
4. Test the System:
   • Run the Python script to test the whole system:

     python classify.py

---

Contributions

Contribution section are still a work in progress. If you would like to contribute or have any suggestions, hit me up with a private message on Discord:

Discord Details:

• Username: .adriancc
• Display Name: Yang
• User ID: 531056443940536320

---

License

This project is licensed under the MIT License.