Drone Swarm Search Environment (DSSE)

Welcome to the official GitHub repository for the Drone Swarm Search Environment (DSSE). This project offers a comprehensive simulation platform designed for developing, testing, and refining search strategies using drone swarms. Researchers and developers will find a versatile toolset supporting a broad spectrum of simulations, which facilitates the exploration of complex drone behaviors and interactions in dynamic, real-world scenarios.

In this repository, we have implemented two distinct types of environments. The first is a dynamic environment that simulates maritime search and rescue operations for shipwreck survivors. It models the movement of individuals in the sea using a dynamic probability matrix, with the objective for drones being to locate and identify these individuals. The second is a environment utilizing the Lagrangian particle simulation from the open-source Opendrift library, which incorporates real-world ocean and wind data to create a probability matrix for drone SAR tasks. In this scenario, drones are tasked with covering the full search area within the lowest time possible, while prioritizing higher probability areas.

📚 Documentation Links

• Documentation Site: Access comprehensive documentation including tutorials, and usage examples for the Drone Swarm Search Environment (DSSE). Ideal for users seeking detailed information about the project's capabilities and how to integrate them into their own applications.

• Algorithm Details: Explore in-depth discussions and source code for the algorithms powering the DSSE. This section is perfect for developers interested in the technical underpinnings and enhancements of the search algorithms.

• PyPI Repository: Visit the PyPI page for DSSE to download the latest release, view release histories, and read additional installation instructions.

DSSE - Search Environment

🎥 Visual Demonstrations

Above: A simulation showing how drones adjust their search pattern over a grid.

🎯 Outcome

| If target is found | If target is not found | :-------------------------:|:-------------------------: | | |

⚡ Quick Start

⚙️ Installation

Quickly install DSSE using pip: bash pip install DSSE `

🛠️ Basic Env Search Usage

```python from DSSE import DroneSwarmSearch

env = DroneSwarmSearch( gridsize=40, rendermode="human", rendergrid=True, rendergradient=True, vector=(1, 1), timesteplimit=300, personamount=4, dispersioninc=0.05, personinitialposition=(15, 15), droneamount=2, dronespeed=10, probabilityofdetection=0.9, prerender_time=0, )

def randompolicy(obs, agents): actions = {} for agent in agents: actions[agent] = env.actionspace(agent).sample() return actions

opt = { "dronespositions": [(10, 5), (10, 10)], "personpod_multipliers": [0.1, 0.4, 0.5, 1.2], "vector": (0.3, 0.3), } observations, info = env.reset(options=opt)

rewards = 0 done = False while not done: actions = randompolicy(observations, env.getagents()) observations, rewards, terminations, truncations, infos = env.step(actions) done = any(terminations.values()) or any(truncations.values())

```

DSSE - Coverage Environment

🎥 Visual Demonstrations

Above: A simulation showing how drones adjust their search pattern over a grid.

⚡ Quick Start

⚙️ Installation

Install DSSE with coverage support using pip: bash pip install DSSE[coverage] `

🛠️ Basic Coverage Usage

```python from DSSE import CoverageDroneSwarmSearch

env = CoverageDroneSwarmSearch( droneamount=3, rendermode="human", disasterposition=(-24.04, -46.17), # (lat, long) prerender_time=10, # hours to simulate )

opt = { "drones_positions": [(0, 10), (10, 10), (20, 10)], } obs, info = env.reset(options=opt)

step = 0 while env.agents: step += 1 actions = {agent: env.action_space(agent).sample() for agent in env.agents} observations, rewards, terminations, truncations, infos = env.step(actions)

print(infos["drone0"]) ```

🤝 Contributing

We welcome contributions from developers to improve and expand our repository. Here are some ways you can contribute:

1. Creating Issues: If you encounter any bugs, have suggestions for new features, or have a question, please create an issue on our GitHub repository. This helps us keep track of what needs to be addressed and prioritize improvements.

2. Submitting Pull Requests (PRs): We encourage you to fork the repository and make your own modifications. Once you have made changes, submit a pull request for review. Ensure your PR includes a clear description of the changes and any relevant information to help us understand the modifications.

Testing Your Contributions

To maintain code stability, we have a suite of tests that must be run before any code is merged. We use Pytest for testing. Before submitting your pull request, make sure to run these tests to ensure that your changes do not introduce any new issues.

To run the tests, use the following command:

bash pytest DSSE/tests/

Our test suite is divided into several parts, each serving a specific purpose:

• Environment Testing: Found in DSSE/tests/test_env.py and DSSE/tests/test_env_coverage.py, these tests ensure that both the search and coverage environments are set up correctly and function as expected. This includes validating the initialization, state updates, and interaction mechanisms for both environments.

• Matrix Testing: Contained in DSSE/tests/test_matrix.py, these tests validate the correctness and functionality of the probability matrix.

📖 How to cite this work

If you use this package, please consider citing it with this piece of BibTeX:

@software{Laffranchi_Falcao_DSSE_An_environment_2024, author = { Laffranchi Falcão, Renato and Custódio Campos de Oliveira, Jorás and Britto Aragão Andrade, Pedro Henrique and Ribeiro Rodrigues, Ricardo and Jailson Barth, Fabrício and Basso Brancalion, José Fernando }, doi = {10.5281/zenodo.12659848}, title = {{DSSE: An environment for simulation of reinforcement learning-empowered drone swarm maritime search and rescue missions}}, url = {https://doi.org/10.5281/zenodo.12659848}, version = {0.2.5}, month = jul, year = {2024} }