Hardware Setup

⚠️ https://www.waveshare.com/wiki/AlphaBot2-Pi#Hardware_setup ⚠️

AlphaPi Zero Agent

A SPADE-based XMPP agent for controlling an AlphaBot2 robot using the Raspberry Pi Zero.

Overview

This project implements a SPADE (Smart Python Agent Development Environment) agent that controls an AlphaBot2 robot. The agent uses XMPP (Extensible Messaging and Presence Protocol) for communication, allowing remote control of the robot through messaging.

Features

• Remote control of AlphaBot2 robot via XMPP messages
• Movement commands: forward, backward, left, right, stop
• Direct motor control with specified speed values
• Containerized deployment with Docker

Requirements

Hardware

• Raspberry Pi Zero (or compatible)
• AlphaBot2 robot kit
• GPIO access

Software

• Python 3.9+
• SPADE 3.3.3
• RPi.GPIO
• spidev
• rpi_ws281x

Installation

Using Docker (Recommended)

0. Install Docker:

   curl -sSL https://get.docker.com | sh
   sudo usermod -aG docker $USER
   restart terminal
   

1. Clone this repository:

   git clone <repository-url>
   cd AlphaPiZeroAgent
   

2. Configure the XMPP settings in docker-compose.yml: (Leave as it is unless configured)

   environment:
     XMPP_SERVER: "prosody"
     XMPP_PORT: 5222
     XMPP_DOMAIN: "prosody"
     XMPP_USERNAME: "alpha-pi-zero-agent"
     XMPP_PASSWORD: "top_secret"

3. Build and run with Docker Compose:

   docker-compose up -d
   

Manual Installation

1. Install the required dependencies:

   pip install -r requirements.txt
   

2. Set the required environment variables:

   export XMPP_SERVER="prosody"
   export XMPP_PORT=5222
   export XMPP_DOMAIN="prosody"
   export XMPP_USERNAME="alpha-pi-zero-agent"
   export XMPP_PASSWORD="top_secret"
   

3. Run the agent:

   python -m agent
   

Usage

Send XMPP messages to the agent with the following commands:

• forward - Move the robot forward for 2 seconds
• backward - Move the robot backward for 2 seconds
• left - Turn the robot left for 2 seconds
• right - Turn the robot right for 2 seconds
• stop - Stop all motors
• motor <left_speed> <right_speed> - Set specific motor speeds (range: -100 to 100)

Project Structure

• agent/__main__.py - Main agent implementation
• agent/alphabotlib/ - Library for controlling the AlphaBot2 hardware
  • AlphaBot2.py - Core motor control functions
  • Additional sensor and control modules

Configuration

The agent uses environment variables for configuration:

• XMPP_SERVER - XMPP server hostname
• XMPP_PORT - XMPP server port (default: 5222)
• XMPP_DOMAIN - XMPP domain
• XMPP_USERNAME - XMPP username for the agent
• XMPP_PASSWORD - XMPP password for the agent

Adding API runner

sudo nano /etc/systemd/system/camera-api.service

[Unit] Description=Camera Flask API After=network.target

[Service] ExecStart=/usr/bin/python3 /home/test/RobotController/agent/camera_api.py WorkingDirectory=/home/test/RobotController/agent Restart=always User=test Environment=PYTHONUNBUFFERED=1

[Install] WantedBy=multi-user.target

sudo systemctl daemon-reexec # Réinitialise systemd (utile après update) sudo systemctl daemon-reload # Recharge les services sudo systemctl enable camera-api # Démarrage auto au boot sudo systemctl start camera-api # Lance maintenant

How to use CameraHandler

The CameraHandler class provides an easy-to-use interface for interacting with the Raspberry Pi camera module. Below are the steps to use it:

Initialization

To use the CameraHandler, you need to initialize the camera and configure it with the desired resolution.

Example Usage

Here’s an example of how to use the CameraHandler class:

from camera_utils import CameraHandler

# Initialize the camera
cam = CameraHandler(resolution=(640, 480))
cam.initialize_camera()

# Capture an image
image = cam.capture_image()

# Save the captured image
cam.save_image(image, output_path="captured_image.jpg")

# Close the camera
cam.close()

Methods Overview

1. initialize_camera()

   • Initializes the camera and configures it with the specified resolution.
   • Raises an error if no camera is detected.

2. capture_image(convert_rgb=True)

   • Captures an image from the camera.
   • Converts the image to RGB format if convert_rgb is set to True.
   • Returns the captured image as a NumPy array.

3. save_image(frame, output_path="captured_image.jpg")

   • Saves the captured image to the specified file path.
   • Default file path: captured_image.jpg.

4. close()

   • Stops the camera and releases resources.

Running the Script Directly

You can also run the camera_api.py script directly to capture and save an image. The script will:

1. Initialize the camera.
2. Capture an image.
3. Save the image as captured_image.jpg.
4. Close the camera.

Run the script using:

python camera_api.py

The captured image will be saved in the same directory as camera_api.py.

Notes

• Ensure the Raspberry Pi camera module is properly connected and enabled in the Raspberry Pi configuration.
• Install the required dependencies (picamera2, opencv-python) before using the script:

  pip install picamera2 opencv-python

License

[Add your license information here]

Acknowledgments

• This project uses the SPADE framework for agent development
• AlphaBot2 is a product of Waveshare