Drone Simulation

A Python-based simulation of drones detecting and processing particles in an ocean environment.

Overview

This project simulates a fleet of drones that fly in various patterns to detect particles in an ocean environment. A catching system follows the drones and processes the detected particles. The simulation visualizes the movement of drones, the catching system, and the particle density in the ocean.

Features

• Multiple Drone Types:

  • Circular Drones: Fly in circular patterns in front of the catching system
  • Lawnmower Drones: Follow a back-and-forth pattern to cover a large area
  • AI Drones: Employ smart dynamic pathing to discover the most particles

• Smart Catching System:

  • Navigates toward high-density areas detected by drones
  • Processes particles when drones are in range
  • Limited turning capability (45° per 3 hours)
  • Constant movement speed (2.78 km/h)

• Dynamic Ocean Environment:

  • Particle clusters that drift with wind patterns
  • Realistic particle distribution with high-density areas
  • Particles are removed from the map when processed by the catching system

• Visualization:

  • Animated GIF output showing the simulation progress
  • Heatmap of particle density
  • Drone and catching system trajectories

Getting Started

Prerequisites

• Python 3.6+
• Required packages: numpy, matplotlib

Installation

1. Clone the repository:

   git clone https://github.com/senster/dronesim.git
   cd dronesim
   

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install dependencies:

   pip install -r requirements.txt
   

Running the Simulation

Run the simulation with circular pattern drones:

python main.py circular

Run the simulation with lawnmower pattern drones:

python main.py lawnmower

Run the simulation with AI drones that use smart dynamic pathing:

python main.py ai

Scanning Strategies

The lawnmower pattern supports different scanning strategies that affect how drones move and scan the area. To list all available strategies:

python main.py --list-strategies

To run the simulation with a specific strategy:

python main.py lawnmower --strategy "1:5 Ratio"

or using the short option:

python main.py lawnmower -s "1:5 Ratio"

If no strategy is specified, a default strategy will be used.

Reproducible Simulations

You can specify a random seed to ensure reproducible particle dispersion across multiple runs:

python main.py lawnmower --seed 12345

This can be combined with strategy selection:

python main.py lawnmower -s "1:5 Ratio" --seed 12345

If no seed is specified, a random seed will be generated and displayed in the output. The seed is also included in the output GIF filename for reference.

Output

The simulation generates an animated GIF in the output directory showing:

• Drone positions and movements (blue)
• Catching system position and movement (yellow)
• Particle density (heatmap)
• Areas where particles have been processed (visible as cleared areas in the heatmap)

Project Structure

• main.py: Entry point for the simulation
• actor.py: Base class for all objects in the simulation
• drone.py: Base class for drone objects
• circular_drone.py: Implementation of drones flying in circular patterns
• lawnmower_drone.py: Implementation of drones flying in lawnmower patterns
• catching_system.py: Implementation of the system that processes particles
• ocean_map.py: Representation of the ocean environment with particle distribution
• simulation_engine.py: Coordinates all actors and runs the simulation
• visualization.py: Handles visualization and animation of the simulation