🚁 Intelligent Delivery Drone Planner

Using Simulated Annealing and Perceptron Classification

👥 Authors

• Muath Hassoun – Computer Engineer
• Zeina Hanani – Computer Engineer

---

🎯 Objective

This project simulates an intelligent drone-based delivery system that finds the most cost-effective delivery route using Simulated Annealing, while avoiding locations with unsafe weather conditions predicted by a Perceptron classifier. Unsafe cities increase the delivery cost due to risk factors.

---

📚 Learning Outcomes

By completing this project, you will learn how to:

• Train and apply a Perceptron classifier for binary classification.
• Integrate classification output into decision-making algorithms.
• Implement Simulated Annealing to solve real-world optimization problems.
• Combine Machine Learning with path planning in delivery systems.
• Build an interactive Graphical User Interface (GUI) for input and visualization.

---

📁 Dataset Description

The project uses an Excel file: weather_data_linearly_separable.xlsx

• Samples: 200
• Features:
  • Temperature (°C)
  • Humidity (%)
  • Wind Speed (km/h)
• Label:
  • SafeToFly:
    • 0 → Safe to fly
    • 1 → Unsafe to fly

---

🛠️ Project Tasks

1. Perceptron Classifier

• Train a Perceptron model using the provided dataset.
• Predict SafeToFly label for new weather data.

2. Delivery Nodes Simulation

• Generate N delivery cities.
• Assign each city:
  • Random or user-defined weather values.
  • 2D coordinates (X, Y).
• Use the Perceptron to predict if each city is safe or unsafe.

3. Cost Matrix Construction

• Calculate Euclidean distance between all city pairs.
• Add a penalty cost (e.g., +50) for visiting cities predicted as unsafe.

4. Simulated Annealing Optimization

• Generate a random initial route.
• Use Simulated Annealing to minimize the total delivery cost:
  • Includes travel distance + unsafe location penalties.

---

🖥️ GUI Features

1. User Input

Allow the user to input:

• Number of delivery cities
• Each city's:
  • Coordinates (X, Y)
  • Weather conditions:
    • Temperature
    • Humidity
    • Wind Speed
• Simulated Annealing parameters:
  • Initial temperature
  • Cooling rate

2. Route Visualization

• Show the initial route on a 2D plane with labels and total cost.
• Show the optimized route after Simulated Annealing:
  • Highlight changes with different colors or line styles.
  • Display the new total route cost.

3. Weather Predictions Display

• For each city, display:
  • ✅ Safe (Green / Checkmark)
  • ⚠️ Unsafe (Red / Warning Icon)
• Optionally display weather data in a table or popup.

4. Cost Analysis

Display a comparison between:

• Initial route cost (distance + unsafe penalties)
• Optimized route cost

---