Hydentify is a tool for parameter identification in multiaffine dynamical systems. Hydentify extends the approach implemented in RoVerGeNe by using a more precise abstraction (to a hybrid system instead of a discrete system).

The approach was presented in the paper Abstraction-Based Parameter Synthesis for Multiaffine Systems at Haifa Verification Conference (HVC) 2015. See below for how to cite this work.

You need to have Matlab, PPL, and GMP installed.

Just extract the file Hydentify/spaceex.zip.001 to the same folder.

Not necessary, we deploy the files in the archive.

Get the latest version from here.

For self-containment, we provide version 2.5.1 here. To install this version, add the folder NuSMV-2.5.1/bin to the PATH variable.

In folder PPLmex/matlab/ open compile.sh and insert the correct name of the Matlab version in the first line (an example is already present). Run compile.sh.

Open Matlab. Move to the Hydentify folder. Run hydentify.

Open functions/get_model_hydentify.m (e.g., in the Matlab editor). Search for model_name =. Enter the name of the model of interest.

Alternatively, there are run scripts in the folder functions/benchmarks/.

@inproceedings{BogomolovSBBKG15,
  author    = {Sergiy Bogomolov and
               Christian Schilling and
               Ezio Bartocci and
               Gr{\'{e}}gory Batt and
               Hui Kong and
               Radu Grosu},
  editor    = {Nir Piterman},
  title     = {Abstraction-Based Parameter Synthesis for Multiaffine Systems},
  booktitle = {{HVC}},
  series    = {LNCS},
  volume    = {9434},
  pages     = {19--35},
  publisher = {Springer},
  year      = {2015},
  url       = {https://doi.org/10.1007/978-3-319-26287-1\_2},
  doi       = {10.1007/978-3-319-26287-1\_2}
}