Overview

rfid-cocktail-bot is a smart, interactive cocktail ordering system that integrates RFID token-based ingredient selection, an ESP32 based Ordering and Pick-Up Station, a JavaScript server, and a UR cobot for automated drink delivery. The system logic is orchestrated via the CPEE process engine.

System Architecture

• RFID Tokens: Represent individual drink ingredients.
• Ordering Station (Input): Shows selected ingredients, confirmed order number, and the status of drink placements.
• Pick-Up Station (Output): Shows the free output spaces, the order number, and placement.
• UR Cobot: Executes drink delivery actions.
• JavaScript Server: Implements and handles CPEE process endpoints, acting as a communication bridge between the CPEE engine and the hardware interface layer.
• CPEE Process Engine: Manages the process workflow and communicates with the server via HTTP requests.

Process Workflow

The drink-serving logic is orchestrated by the CPEE process model:

Main Steps:

1. Refresh the Server: Clears all variables stored from the previous run.

branch 1

1. Wait for Orders:
   Sends a CALLBACK URL to the server and waits.
   When a new order is submitted by a user, the server responds with an UPDATE, pushing the order into the order_list array in CPEE.

branch 2

1. Check Order List:
   Continuously loops to check if there is at least one pending order.
2. Fulfill First Order:
   Retrieves the first order from the order_list and assigns it to the serve_order variable.
3. Timeout (Preparation Simulation):
   Simulates a drink preparation delay.
4. Find an Available Position
   Sends a request to the server to get glass_x, glass_y, available_position.
   • A fixed 2×5 grid of glasses is used cyclically, so refilling is only needed after every 10 orders.
   • The server determines the available output position based on weight cell readings. If all positions are occupied, it stores the CALLBACK URL and retries every second until one becomes free.
   • The LED display is updated with the assigned output position.
5. Send Coordinates to UR Cobot:
   The CPEE sends glass_x, glass_y, available_position to the UR cobot.
6. Serve Drink:
   The UR cobot picks up the glass from position (glass_x, glass_y) and places it at available_position.
   The task is completed after receiving a response from the UR cobot, indicating that the actions have been successfully executed.
7. Mark Order as Finished:
   Informs the server that the process has been completed.
   • The finished order is moved from order_list to order_history.
   • The serve_order variable is cleared.

The parallel structure ensures continuous order reception while drinks are being prepared and served.

User Interaction Flow

1. Selecting Ingredients:

   • Each RFID token corresponds to a drink ingredient.
   • Toss an RFID token into the reader ring (input) to select an ingredient.
   • Once a token is read, the corresponding ingredient name is displayed on the LED screen.
   • If more than four ingredients are selected, the extra tokens will be ignored.

2. Confirming or Clearing Selection:

   • Red Button: Clears the current ingredient selection.
   • Green Button: Confirms the order.
   • Upon confirmation, a unique order number is generated and displayed on the LED screen.
   • The system then simulates drink preparation with a 10-second timeout.

3. Drink Serving:

   • The UR cobot places the finished drink at an available position in the pickup station (output).
   • The LED display shows current order numbers at each position (e.g., 0:101, 1: , 2: , 3: ).
   • Once a user picks up the drink, the corresponding position is cleared (e.g., 0: , 1: , 2: , 3: ).

Example Interaction

1. User drops 3 RFID tokens into the reader (input).
2. LED screen shows: Vodka, Lemon Juice, Soda.
3. User presses the green button to confirm.
4. LED shows: Order number: 102 for 10s.
5. After the drink is prepared and placed by the UR cobot (output), the LED updates the serving station status: 0:102, 1: , 2: , 3:
   (The drink has been placed at position 0).

Setup Instructions

To run the system:

1. Connect all hardware components (RFID reader, LEDs, buttons, weight sensors, etc.).
2. Power on the UR cobot and start the cobot-side server.
3. On the lehre.bpm server:
   SSH into the target machine, then run:

   cd /srv/gruppe/students/ge27zap/endpoint/
   node server.js

4. Access the CPEE Process Model:
   Launch the Main_process via the following path: /Teaching/Prak/TUM-Prak-24-WS/Jingyi Tang/restructured