Skip to content

Latest commit

 

History

History
211 lines (158 loc) · 5.61 KB

README.md

File metadata and controls

211 lines (158 loc) · 5.61 KB
Raw

milp_mespp

Overview | Code Structure | Installation | Examples | Troubleshooting | Citing this work

Overview

MILP models for the Multi-robot Efficient Search Path Planning (MESPP) problem: team of robots is deployed in a graph-represented environment to capture a moving target within a given deadline. Supports multiple searchers, arbitrary capture ranges, and false negatives simultaneously.

Citing this work

If you use this algorithm or code, please cite our paper:

Thank you!

@article{asfora2020,
  author={Asfora, Beatriz A. and Banfi, Jacopo and Campbell, Mark},
  journal={IEEE Robotics and Automation Letters}, 
  title={Mixed-Integer Linear Programming Models for Multi-Robot Non-Adversarial Search}, 
  year={2020},
  volume={5},
  number={4},
  pages={6805-6812},
  doi={10.1109/LRA.2020.3017473}}

Code Structure

milp_mespp
├── classes
│   ├── belief
│   ├── inputs
│   ├── searcher
│   ├── solver_data
│   └── target
├── core
│   ├── extract_info
│   ├── create_parameters
│   ├── construct_model
│   ├── milp_fun
│   ├── plan_fun
│   ├── sim_fun
│   ├── retrieve_data
│   └── plot_fun
├── data
├── examples
│   ├── plan_only
│   └── numerical_sim
├── graphs
└── tests

Installation Guide

This project was developed in Python 3.6 and updated to Python 3.8. It uses the following Python libraries: datetime, sys, os, pickle, numpy, matplotlib, igraph, gurobipy.

Start by cloning this repository,

git clone https://github.com/basfora/milp_mespp.git

This project may be offered as a Python package in the future, but at the moment the code is still under construction. Please pull for updates every now and then, and report any bugs - I will do my best to fix them.

Installing gurobipy

Gurobi License and installation instructions here

Important: changing default saving location of license file will cause errors! Don't do that.

To install gurobipy, run (change path, OS and version accordingly),

cd path-to-folder/gurobi902/linux64/
sudo python3 setup.py install

If you are using PyCharm, you might need to also run these commands on PyCharm's terminal.

Run install script

This will install all the other necessary Python libraries and add the folder to your Python path system variable. From inside the milp_mespp folder, run on terminal:

chmod +x install_script.sh
./install_script.sh

Press ENTER and enter your user password when prompted.

This script assumes Ubunty 18.04. Last tested version: Ubuntu 22.04, Gurobi 10.00 For other OS the project code should work, but you will need to install the libraries/set path manually.

When the installation is done, you should see this on your terminal (actual vertex numbers may vary):

--
Planned path: 
Searcher 1: [27, 54, 53, 58, 44, 45, 46, 47, 47, 47, 48]
t = 0
Target vertex: 13
Searcher 1: vertex 27 

--
Time step 1 
--
t = 1
Target vertex: 13
Searcher 1: vertex 54

. . .

--
Time step 9 
--
t = 9
Target vertex: 44
Searcher 1: vertex 47 

--
Time step 10
--
t = 10
Target vertex: 49
Searcher 1: vertex 48

This means both the planner and simulator are working.

Source your .bashrc file, by running on terminal:

source ~/.bashrc

Your installation is now complete.

Examples

To learn how to change specs and run multiple instances, check examples/numerical_sim.

Data and plots will be saved in a milp_mespp/data folder (created the first time you run numerical_sim.py).

Troubleshooting

If you try to run the install_script.sh and get the error bash: ./install_script.sh: Permission denied, make sure file Properties > Permissions > Execute: Allow executing file as program is checked

Manual install

If you don't want to use the install_script or run into errors, you can install things manually.

Installing packaging tools

On terminal

sudo apt-get install python3-distutils
sudo apt-get install python3-apt
sudo apt install python3-pip

Installing commonly used libraries

On terminal:

python3 -m pip install -U matplotlib
python3 -m pip install -U numpy
python3 -m pip install -U pytest
sudo apt-get install build-essential
sudo apt-get install python3.8-dev

Installing igraph

sudo add-apt-repository ppa:igraph/ppa
sudo apt-get update
sudo apt-get install python-igraph

If it throws errors, run:

sudo apt-get install bison flex
python3 -m pip install python-igraph

Setting path

Add folder absolute path to your $PYTHONPATH system variable. On Linux OS, paste this on your .bashrc file (change path accordingly):

export PYTHONPATH="${PYTHONPATH}:path-to-folder/milp_mespp"

Don't forget to source it (or restart your computer):

source ~/.bashrc

Run default simulator

Make sure things are working by running the simulator with default values.

cd milp_mespp/core
python3 sim_fun.py

You should see the same output as with the install script (see above).

Author

Beatriz Asfora

Acknowledgements

Dr. Jacopo Banfi
Prof. Mark Campbell