plafosim 0.17.2

A simple and scalable simulator for platoon formation.

PlaFoSim - A simple and scalable simulator for platoon formation

PlaFoSim aims to facilitate and accelerate the research of platoon maneuvers and formation for individually driven vehicles. While the main focus of the simulator is on the assignment process, simulation of advertisements and maneuvers is implemented in a more abstract way.

A conceptional overview of the process of Platoon Formation that is employed in PlaFoSim [1-2]:

Scenario	Advertisement	Assignment	Maneuver
A new vehicle enters the highway.	The vehicle advertises itself as interested in Platooning.	A Vehicle-to-Platoon assignment is computed.	The new vehicle performs a join maneuver.

PlaFoSim has been published at IEEE VNC 2021:

Julian Heinovski, Dominik S. Buse and Falko Dressler, "Scalable Simulation of Platoon Formation Maneuvers with PlaFoSim," Proceedings of 13th IEEE Vehicular Networking Conference (VNC 2021), Poster Session, Virtual Conference, November 2021, pp. 137–138.

NOTE: PlaFoSim is (still) under heavy development.

---

Installation

• Install Python (>=3.7,<=3.9)
• Install PlaFoSim from pypi: pip install plafosim

NOTE: The project is developed and tested only on Linux.

Running a Simulation

NOTE: Since PlaFoSim is a command-line interface (CLI) application, running PlaFoSim requires using a shell.

Quickstart

Run PlaFoSim with:

plafosim

You can use PlaFoSim's help to get a list of available parameters:

plafosim -h, --help

Or, run a simulation with the default configuration (-d):

plafosim -d

Live GUI

You can visualize the simulation via a simple live GUI based on sumo-gui, using the argument gui:

plafosim --gui

A screenshot of PlaFoSim's live GUI showing 2 platoons and various individual vehicles. Copyright © 2021 IEEE.

More options for the live GUI can be found within the GUI properties section of the help.

NOTE: This requires installation of SUMO (>=1.6.0) and declaration of the SUMO_HOME variable (see documentation).

Advanced Simulation Control

You can use a variety of different parameters to customize the scenario and the simulation itself. E.g., use the parameter vehicles to configure the number of vehicles in the simulation:

plafosim --vehicles 1000

The available parameters are grouped into different categories:

- road network properties
- vehicle properties
- trip properties
- communication properties
- platoon properties
- formation properties
- infrastructure properties
- simulation properties
- GUI properties
- result recording properties

You can find a list of available parameters for each category in the help:

plafosim -h, --help

Examples

# Configure a 100km freeway with ramps at every 10km
plafosim --road-length 100 --ramp-interval 10

# Configure random (normally distributed) desired driving speed of 130km/h
plafosim --random-desired-speed true --desired-speed 36

# Configure random trips for 500 vehicles
plafosim --vehicles 500 --random-depart-position true --random-arrival-position true --depart-desired true

# Pre fill the freeway with 1000 vehicles
plafosim --vehicles 1000 --pre-fill true

# Configure 50% of the vehicles with Advanced Cruise Control (ACC) and a headway time of 1.5s
plafosim --penetration 0.5 --acc-headway-time 1.5

# Enable a simple, distributed platoon formation algorithm [1] in order to form platoons every 30s
plafosim --formation-algorithm SpeedPosition --formation-strategy distributed --execution-interval 30

Faster Simulation

You can speed up the simulation performance by enabling Python's optimization PYTHONOPTIMIZE, e.g., in order to disable assertions:

PYTHONOPTIMIZE=1 plafosim

See the Python documention for more details.

Re-Playing a Simulation

The simulation can write a trace file including the mobility details of every simulated vehicle (default results_vehicle_traces.csv). You can replay the simulation in the GUI (see above) based on the trace file by using the corresponding binary:

plafosim-replay results_vehicle_traces.csv

To see all options of this script, run:

plafosim-replay -h, --help

NOTE: This requires installation of SUMO (>=1.6.0) and declaration of the SUMO_HOME variable.

Recording of Screenshots and Video from a Simulation

PlaFoSim offers functionality to automatically record a screenshot of the GUI in every simulation step with

plafosim --gui --screenshot-file screenshot

These screenshots can be used to create a video by using an integrated script based on ffmpeg:

plafosim-img2video 'screenshot_*.png' video.mp4

NOTE: This requires installation of ffmpeg.

Documentation

Documentation is available via this README and our paper. Further, technical documentation is available via python docstrings and comments in the code.

NOTE: A proper documentation is still under development.

Extending PlaFoSim

In order to extend PlaFoSim, you first need ot install it from source.

• Install poetry: pip install poetry
• Clone this repository: git clone https://github.com/heinovski/plafosim.git
• Navigate to newly created directory of the cloned repository in the command-line
• Install PlaFoSim from source in editable mode: poetry install
• Run PlaFoSim in the virtual environment: poetry run plafosim You can also activate the virtual enviroment first with poetry shell and run the commands as usual (see above).

Adding a new Formation Algorithm

In order to add a new formation algorithm, you need to follow these steps:

• Create a new sub-class of FormationAlgorithm (see formation_algorithm.py) within the sub-directory algorithms. You can use the Dummy algorithm (see algorithms/dummy.py) as an example.
• Add specific arguments for your algorithm to the argument parser group within the new sub-class if necessary.

You should now be able to use your new algorithm with

plafosim --formation-algorithm dummy_algorithm_name

Contributing to the Project

In order to contribute, please follow these steps:

• Install PlaFoSim from source (see Extending PlaFoSim)
• Make desired changes and adjust the documentation if required
• Run the tests located in tests as well as the validation scripts located in scripts (see .drone.yml for details)
• Submit a Pull Request (PR)

Documenting

When making changes to the code, make sure to add or adjust corresponding documentation in form of python docstrings. Those should use the Numpy docstring format for Sphinx and follow the style guide by numpydoc to ensure consistency and compatibility. See also the style guide by pandas.

Testing

When adding methods and functions, make sure to add corresponding unit tests for py.test. The tests are located under tests and can be executed with ./scripts/run-pytest.sh. This will also generate a test coverage report.

Validation

To validate the behavior of PlaFoSim, it is compared to SUMO 1.6.0 by means of simulation results (e.g., vehicle traces). The corresponding scripts are located under scripts and executed withn CI/CD pipelines. You can have a look at .drone.yml for details regarding the execution.

Profiling

You can profile the runtime of PlaFoSim's code by using cProfile:

poetry run python -m cProfile -o profile.out -m plafosim.cli.plafosim

You can visualize the results of the profiling run by using SnakeViz:

snakeviz profile.out

Contributors & Citing

PlaFoSim was originally designed and built by Julian Heinovski with the help of Dominik S. Buse. It is currently maintained by Julian Heinovski. The list of all authors can be found here.

If you are working with PlaFoSim, we would appreciate a citation of our paper:

Julian Heinovski, Dominik S. Buse and Falko Dressler, "Scalable Simulation of Platoon Formation Maneuvers with PlaFoSim," Proceedings of 13th IEEE Vehicular Networking Conference (VNC 2021), Poster Session, Virtual Conference, November 2021, pp. 137–138.

@inproceedings{heinovski2021scalable,
    author = {Heinovski, Julian and Buse, Dominik S. and Dressler, Falko},
    doi = {10.1109/VNC52810.2021.9644678},
    title = {{Scalable Simulation of Platoon Formation Maneuvers with PlaFoSim}},
    pages = {137--138},
    publisher = {IEEE},
    issn = {2157-9865},
    isbn = {978-1-66544-450-7},
    address = {Virtual Conference},
    booktitle = {13th IEEE Vehicular Networking Conference (VNC 2021), Poster Session},
    month = {11},
    year = {2021},
}

License

PlaFoSim is licensed under the terms of the GNU General Public License 3.0 or later.

# Copyright (c) 2020-2023 Julian Heinovski <[e-mail]>
#
# SPDX-License-Identifier: GPL-3.0-or-later
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <https://www.gnu.org/licenses/>.

List of References

[1] Julian Heinovski and Falko Dressler, "Platoon Formation: Optimized Car to Platoon Assignment Strategies and Protocols," Proceedings of 10th IEEE Vehicular Networking Conference (VNC 2018), Taipei, Taiwan, December 2018.

[2] Julian Heinovski, "Platoon Formation: Car-to-Platoon Assignments for Individual Cars," Proceedings of International Conference on Networked Systems (NetSys 2019), PhD Forum, Munich, Germany, March 2019.