This project integrates computer vision with hardware control to detect fire and respond with coordinated motor actions. The system:

• Uses OpenCV to detect fire based on HSV color thresholds.
• Adjusts the pitch (e.g., water nozzle angle) using a stepper motor.
• Pivots the robot chassis by running one wheel at a time (two-wheel drive).
• Activates a water pump (via a separate DC motor HAT on I²C address 0xA0) only if a stable fire is detected for at least 4 seconds.

• Computer Vision (OpenCV): Detects fire regions by analyzing video frames.
• Stepper Motor: Adjusts the pitch based on the vertical position of the detected fire.
• DC Motors (Wheels): Control the chassis; each motor powers one wheel so that the robot can pivot.
• Water Pump: Controlled by a separate DC motor HAT and activated when the fire remains stable.

• Imports:
  The code uses:

  • OpenCV and NumPy for image processing.
  • time for time tracking.
  • board, MotorKit, and stepper (from Adafruit) for hardware control.

• MotorKit Instances:

  • Primary MotorKit (kit): Controls the wheels (DC motors) and the stepper motor for pitch adjustment.
  • Pump MotorKit (pump_kit): Initialized with I²C address 0xA0, dedicated to controlling the water pump.

• spin_one_wheel:
  Runs a single wheel motor for a specified duration.

  • Parameters:
    • wheel: 1 for the left wheel or 2 for the right wheel.
    • direction: 'forward' or 'backward'.
    • duration: Time in seconds to run the motor.
    • throttle: Motor power (0.0 to 1.0).
  • Behavior:
    Sets the motor’s throttle (positive for forward, negative for backward), waits, and then stops the motor.

• spin_wheels_sequence:
  Activates each wheel individually in sequence.

  • Running one wheel at a time causes the robot to pivot, which helps spread the water more effectively.

• activate_water_pump:
  Activates the water pump via the pump HAT.
  • Parameters:
    • duration: Seconds to run the pump.
    • throttle: Pump power (0.0 to 1.0).
  • Behavior:
    Sets the pump motor’s throttle, runs it for the specified duration, and then turns it off.

• Video Capture:
  Opens the video file (./software/fire7.mov) for processing.

• Preprocessing:

  • Converts each frame from BGR to HSV.
  • Applies Gaussian blur to reduce noise.

• Thresholding:

  • Defines an HSV range (with lower_fire and upper_fire) to isolate low-saturation, very bright fire-like regions.
  • Creates a binary mask for these regions.

• Morphological Operations:
  Cleans the mask using opening and closing to remove noise.

• Contour Detection:

  • Extracts contours from the mask.
  • For contours with an area greater than 150 pixels:
    • Calculates the bounding box, center coordinates, and the brightest point (using the V channel).
    • Draws these details on the frame and prints them to the console.

• Purpose:
  Adjust the pitch of the water nozzle (or camera) based on the vertical position of the detected fire.

• Implementation:

  • Computes the average y-coordinate of the detected fire regions.
  • Compares it to the frame’s vertical center.
  • If the fire is above the center (beyond a small tolerance), the stepper motor steps forward (pitch up). If below, it steps backward (pitch down).

• Purpose:
  To pivot the robot so that water can be spread more efficiently over the fire.

• Implementation:

  • If fire is detected and a minimum interval (e.g., 5 seconds) has passed, the spin_wheels_sequence() function is called.
  • This function activates each wheel sequentially, causing a pivoting motion.

• Stability Tracking:

  • Computes the average fire center (both x and y coordinates) from the detected regions.
  • Uses a pixel tolerance (20 pixels) to determine if the fire’s position is stable between frames.
  • If no previous center is recorded, the current center and time are saved.

• Stability Check and Activation:

  • If the fire remains within the tolerance for at least 4 seconds, and if a cooldown period (10 seconds) has passed since the last activation, the water pump is activated for 2 seconds at full throttle.
  • If the fire moves significantly, the stability timer resets.

• Frame Processing:
  The script processes each video frame, running fire detection, adjusting the pitch, pivoting the chassis, and managing the water pump activation.

• Display and Exit:
  The processed frame is shown in a window, and the loop continues until the user presses q.

• Cleanup:
  Once the loop exits, the video capture is released and all OpenCV windows are closed.

1. Hardware Setup:

   • Connect the primary MotorKit for wheels and the stepper motor to the default I²C address.
   • Connect the water pump HAT to I²C address 0xA0 (adjust if necessary).
   • Place the video file fire7.mov in the ./software/ directory (or update the path in the code).

2. Install Dependencies:

   pip install opencv-python numpy adafruit-circuitpython-motorkit adafruit-circuitpython-motor