Tipsy the AI Cocktail Maker Project

This project is a multi-part cocktail maker system designed to run on a Raspberry Pi. It integrates three separate modules:

1. Streamlit App (app.py):
   Allows users to configure pump settings, generate cocktail recipes (using an OpenAI API), and view a recipe book with cocktail images and details.

2. Pygame Interface (interface.py):
   Provides a full-screen, swipeable interface for cocktail selection. The interface displays a background image (tipsy.png), overlays the selected cocktail logo, and includes extra interactive logos (single.png and double.png) to select drink mode (single or double). It features various animations such as text zoom and logo pop effects, and displays pouring/loading overlays.

3. Pump Controller (controller.py):
   Reads the selected cocktail and pump configuration (from cocktails.json and pump_config.json), then controls 12 pumps via the Raspberry Pi GPIO (using L91105 motor drivers) to mix the drink. The controller uses a mapping of ingredients to pump pins.

4. Main Launcher (main.py) (optional):
   A helper script to launch both the Streamlit app and Pygame interface concurrently.

---

Features

• Cocktail Recipe Generation:
  Generates cocktail recipes (via OpenAI API) based on pump configuration and bartender requests.

• Swipe & Mode Selection Interface:
  Use touch/mouse swipe gestures to navigate cocktail logos. Tap the extra logos (single.png and double.png) to select drink mode, triggering animations and overlays.

• Pump Control:
  Uses Raspberry Pi GPIO and L91105 motor drivers to run pumps based on the selected cocktail’s ingredients.

• Configurable Pump Setup:
  Pump-to-ingredient mapping is stored in pump_config.json.

• Persistent API Key:
  The Streamlit app prompts for an OpenAI API key (if not found in a .env file) and saves it for future use.

---

Hardware Requirements

• Raspberry Pi (any recent model)

• L91105 Motor Drivers for each pump

• 12 Pumps connected to the L91105 boards, with the following GPIO pin configuration:

  Pump	IA Pin	IB Pin
  Pump 1	GPIO17	GPIO4
  Pump 2	GPIO22	GPIO27
  Pump 3	GPIO10	GPIO9
  Pump 4	GPIO5	GPIO11
  Pump 5	GPIO13	GPIO6
  Pump 6	GPIO26	GPIO19
  Pump 7	GPIO20	GPIO21
  Pump 8	GPIO12	GPIO16
  Pump 9	GPIO8	GPIO7
  Pump 10	GPIO24	GPIO25
  Pump 11	GPIO18	GPIO23
  Pump 12	GPIO14	GPIO15

• Touchscreen or Mouse for interacting with the Pygame interface

PCB Files

The project includes custom PCB designs for the pump controller circuitry. These files are located in the /PCB Board Production folder and contain everything needed for fabrication:

• Gerber Files: All layers required for manufacturing the PCB.
• Bill of Materials (BOM): A complete list of components.
• Component Position Files: Placement and assembly details.

You can send these files directly to a PCB manufacturer (such as JLCPCB) to have the custom boards produced.

---

Software Dependencies

• Python 3.x
• Streamlit
  pip install streamlit
• Pygame
  pip install pygame
• RPi.GPIO (or similar, for Raspberry Pi GPIO control)
  pip install RPi.GPIO
• python-dotenv
  pip install python-dotenv
• Requests
  pip install requests
• OpenAI Python Library
  pip install openai
• Pydantic
  pip install pydantic

---

Installation & Setup

1. Clone the Repository:

   git clone https://github.com/Concept-Bytes/Tipsy
   cd Tipsy

2. Install Dependencies:

   pip install -r requirements.txt

   (Alternatively, install the packages listed above individually.)

3. Configure GPIO & Hardware:
   Wire your pumps to the Raspberry Pi as per the pin mapping provided above. Ensure your L91105 drivers and pumps are correctly connected.

4. Set Up API Key:
   When running the Streamlit app (app.py), you will be prompted to enter your OpenAI API key. This key will be saved to a .env file for future use.

5. Prepare JSON Files:

   • Edit pump_config.json to map pumps to your desired ingredients.
   • Generate cocktail recipes via the Streamlit interface (app.py) to create/update cocktails.json.

6. Place Image Assets:

   • Ensure tipsy.png is in the project root.
   • Place your cocktail logo images in the drink_logos folder.
   • Ensure single.png, double.png, pouring.png, and loading.png are in the project root (or adjust paths accordingly).

7. PCB Board Production Files: Check the /PCB Board Production folder for Gerber files, BOM, and component placement files required for PCB manufacturing. Send these files to your chosen PCB manufacturer (e.g., JLCPCB).

---

Running the Project

Option 1: Launch Separately

• Streamlit App:

  streamlit run app.py

• Pygame Interface:

  python interface.py

• Controller:
  After selecting a cocktail and mode, run the controller on the Raspberry Pi:

  python controller.py

Option 2: Launch Together Using Main Launcher

• Main Launcher (main.py):

  python main.py

  This script launches both the Streamlit app and the Pygame interface concurrently. (You can run the controller separately when ready.)

---

Controller Operation

The controller.py script will:

1. Read pump_config.json and cocktails.json.
2. Determine which cocktail is selected by reading selected_cocktail.txt.
3. Determine the drink mode (single or double) by reading selected_mode.txt.
4. Activate the appropriate pumps via GPIO for the required duration based on the cocktail recipe.
   Adjust the conversion factor (seconds per ounce) in controller.py to suit your pump flow rate.

---

Troubleshooting

• GPIO Issues:
  Ensure that your wiring matches the pin mappings in controller.py and that your Raspberry Pi user has permissions to access GPIO.

• Image Loading Errors:
  Verify that all image files (tipsy.png, pouring.png, loading.png, logos, etc.) are present and paths are correct.

• API Key Issues:
  If the API key prompt appears on every run, ensure the .env file is being written to and that you have appropriate write permissions.

• Dependency Errors:
  Check that all Python dependencies are installed correctly. Use pip freeze to verify.

---

License

Include your project license here (e.g., MIT License).

---

Acknowledgments

Mention any libraries, tutorials, or contributors that helped build this project.

---

This README provides an overview of the system, setup instructions, hardware and PCB file details, and troubleshooting tips for your cocktail maker project on a Raspberry Pi.