https://github.com/schillic/ha2stateflow
A converter from a hybrid automaton model to a Stateflow model
https://github.com/schillic/ha2stateflow
hybrid-automaton source-to-source stateflow
Last synced: 4 months ago
JSON representation
A converter from a hybrid automaton model to a Stateflow model
- Host: GitHub
- URL: https://github.com/schillic/ha2stateflow
- Owner: schillic
- Created: 2018-12-06T18:46:07.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2021-11-09T18:45:00.000Z (over 3 years ago)
- Last Synced: 2025-01-12T13:51:57.161Z (6 months ago)
- Topics: hybrid-automaton, source-to-source, stateflow
- Language: C
- Size: 10.6 MB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Citation: CITATION.bib
Awesome Lists containing this project
README
This is supplementary material for the article [*Hybrid automata: from verification to implementation*](http://dx.doi.org/10.1007/s10009-017-0458-1), STTT, 2017.
See [below](#citation) for how to cite this work.
The tool has also been integrated in [HyST](https://github.com/verivital/hyst).
---
The archive contains:
* a folder `examples` that includes all benchmarks
* a folder `lib` that includes an executable Jar file of [Hyst](http://verivital.com/hyst/)
* a folder `src` that includes the source code of the translator
* a folder `buck_interval` that includes all necessary files to reproduce the experimental results for the buck converter evaluated in the paper;
Particularly, this folder contains the following:
* SpaceEx model files for buck converter using intervals
* a folder "buck_interval_data" that includes data obtained from SLSF simulation, reachability analysis with SpaceEx, and hardware experiment with dSPACE DS1103 system
* a folder "code_generation" that includes C code generation for the buck converter by dSPACE
* a file "dSPACE Experiment Report" that describes in details the dSPACE experiment for the buck converter
---
To run the translator, do the following:
* Open Matlab.
* Move to the folder where you extracted the files.
* Add the source code to the Matlab path (in the "Current Folder" view).
For that, right-click on the folder "src" and select "Add to Path" and the subcategory "Selected Folders and Subfolders".
* Run the following command (where you replace "MODEL_NAME" by a meaningful relative path of your model file):
```matlab
SpaceExToStateflow('MODEL_NAME', '-s');
```
This produces a StateFlow model (which might take some time, especially when the Simulink libraries have not been loaded yet).
Example call:
```matlab
SpaceExToStateflow('examples\vanderpol\vanderpol.xml', '-s');
```
---
To run the simulation loop, run the following (parametrized) command:
```matlab
simulationLoop('MODEL_NAME', NUM_SIMULATION, MAX_TIME, NUM_BACKTRACK);
```
Here, the parameters denote the following:
* the model name (automatically taken from the `*.xml` file)
* the number of simulations
* the maximum simulation time
* the number of backtrackings
Again, this might take some time the first time of the call because Simulink compiles the model in the background.
Example call:
```matlab
simulationLoop('vanderpol', 5, 50, 3);
```
---
To reproduce the table of benchmarks in the appendix, run the following script file:
```matlab
generateTable
```
# Citation
```bibtex
@article{BakBBJNS19,
author = {Stanley Bak and
Omar Ali Beg and
Sergiy Bogomolov and
Taylor T. Johnson and
Luan Viet Nguyen and
Christian Schilling},
title = {Hybrid automata: from verification to implementation},
journal = {Int. J. Softw. Tools Technol. Transf.},
volume = {21},
number = {1},
pages = {87--104},
year = {2019},
url = {https://doi.org/10.1007/s10009-017-0458-1},
doi = {10.1007/s10009-017-0458-1}
}
```