Skip to content

Object Detection Subsystem

The Object Detection subsystem is a vision processing system that uses PhotonVision to detect and track game objects on the field in real-time. It maintains lists of both stable (consistently tracked) and pending (newly detected) objects, providing filtered and reliable object detection data to other subsystems.

Constructor

public ObjectDetectionSubsystem(Supplier<Pose2d> robotPose, ObjectDetectionIO... io)

Parameters: - robotPose: A supplier function that provides the current robot pose in field coordinates - io: One or more ObjectDetectionIO interfaces representing the vision cameras

The following commands utilize this subsystem: - DriveFacingBestObject - Automatically rotates the robot to face detected objects while driving - PathfindToBestObject - Autonomously navigates the robot to approach detected objects

Configuration Requirements

The subsystem requires several constants to be configured in the ObjectDetectionConstants class:

  1. Camera Configurations:

    • CAMERA_CONFIGS: Array of camera configuration parameters
    • POSITION_MATCH_TOLERANCE: Distance threshold for matching detected objects
    • HISTORY_LENGTH: Number of frames to keep in detection history
    • REQUIRED_DETECTIONS: Number of detections required for stability
    • MISSING_FRAMES_THRESHOLD: Number of missed frames before losing stability

  2. Game Element Definitions:

    • GAME_ELEMENTS: Array mapping detected class IDs to game elements

  3. PhotonVision Configuration:

    • Each camera must be properly configured in PhotonVision with:
      • Appropriate pipelines for object detection
      • Correct camera mounting position and angle
      • Calibrated camera parameters

Requirements

  • PhotonVision installed and configured on your robot
  • At least one compatible camera connected and configured
  • WPILib 2024 or newer

For detailed method documentation, refer to the JavaDoc reference.