Hybrid vehicle single track model

This repository contains Julia and Python code to reproduce numerical results of the manuscript: Vehicle single track modeling using physics informed neural differential equations

How to run it?

• Install Julia (v1.6.7 was used herein) and VScode with Julia for VScode extension.
• Configure the local environment with instantiate in Julia's built in package manager Pgk. You can switch from julia> to (your path) pkg> by pressing ] and backspace key.
• Open the files in /doc folder in VScode and run them with either Shift + Enter line by line in Julia's REPL or the entire file with Ctrl + F5.

As alternative, you can run the files without VScode from terminal with

julia
]
activate .
<- (delete key)
julia> include("path/to/script.jl")

If this works, you can run the script files:

julia> include("doc/ode.jl")
julia> include("doc/node.jl")
julia> include("doc/hybrid_node.jl")

to reproduce the result figures, which are all stored in results/ folder as .html files. You can whatch these figures with any browser.

How to generate the reference model .mat files?

The reference model is implemented with Python in the reference_model folder. You will find there a readme. However, the file reference_sinlge_track_drift.mat contains the exported time series from the reference model and is already stored in the root folder of this repository.

Purpose

This software is a research prototype, solely developed for and published as part of the publication. It will neither be maintained nor monitored in any way.

License

This software is licensed under AGPL-3.0 license.