Skip to content

README.md

Latest commit

 

History

History
60 lines (49 loc) · 1.51 KB

README.md

File metadata and controls

60 lines (49 loc) · 1.51 KB

🧠 Pickasso Vision Inference & Ethical Decision-Making

This repository contains the Vision Inference and Ethical Decision-Making pipeline for the Pickasso Robot Project.

📚 Project Structure

  • models/: Place your pre-trained YOLO .pt models here.
  • test_images/: Add test images to simulate object detection.
  • vision_ethics_integration.py: Main script performing detection and applying ethical decisions.
  • task_queue.json: Auto-generated task queue based on detected objects.

🚀 How to Run

  1. Clone the Repository:
git clone https://github.com/AapseMatlb/pickasso-vision-inference.git
cd pickasso-vision-inference
  1. Install Dependencies:
pip install torch torchvision pandas
  1. Run the Script:
python3 vision_ethics_integration.py

✅ Ethical Decision Mapping

Item Action
Trash Pick
Dangerous Item Alert
Personal Item Ignore
Stationary Ignore
Human Detected Pause and Ask for Help

📂 Example Output (task_queue.json)

[
    {
        "image": "office1.jpg",
        "item": "trash",
        "confidence": 0.92,
        "action": "Pick"
    },
    {
        "image": "office1.jpg",
        "item": "blade",
        "confidence": 0.88,
        "action": "Alert"
    }
]

For detailed simulation of Navigation, Manipulation, and HMI, refer to the main Pickasso project repository.