/formal-robot-assisted-evacuation

An agent-based evacuation model to study socio-cultural factors influencing the evacuation process

Primary LanguageNetLogoMIT LicenseMIT

Towards Verified Robot-Controlled SaR Scenarios

The repository contains a tool to generate and simulate formally verified strategies for Search-and-Rescue (SaR) robots in evacuation scenarios.

The tool builds upon the work presented by Gavidia-Calderon et al.

Authors:

Name E-mail address
Gavidia-Calderon Carlos cgavidiac@gmail.com
Lestingi Livia livia.lestingi@polimi.it

Project Structure

The tool generates an Uppaal Stratego model based on input parameters.

An Uppaal Stratego experiment is then performed to generate a verified and optimal strategy for the SaR robot.

The generated strategy is then parsed by the controller (invoked at runtime by the simulator) to select the best decision based on the current state of the system.

Configuration File Setup

The tool requires a configuration file whose content must be adjusted to match your environment. Make sure that the following properties match your environment:

  • UPPAAL_PATH is the path to Uppaal verifyta executable

Python Dependencies

Install the required dependencies:

pip install -r $REPO_PATH/requirements.txt

Add the repo path to your Python path (fixes ModuleNotFoundError while trying to execute from command line):

export PYTHONPATH="${PYTHONPATH}:$REPO_PATH"

Initialize Git submodules with:

git submodule init
git submodule update

Copyright © 2024 Carlos Gavidia-Calderon, Livia Lestingi