Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Using Breach for data generation #3

Open
5 tasks
nikos-kekatos opened this issue May 3, 2020 · 0 comments
Open
5 tasks

Using Breach for data generation #3

nikos-kekatos opened this issue May 3, 2020 · 0 comments
Assignees
Labels
enhancement New feature or request

Comments

@nikos-kekatos
Copy link
Owner

nikos-kekatos commented May 3, 2020

Add an extra option to use Breach as an input generation technique for Simulink. Currently, three options are supported:

  1. Random inputs (uniformly distributed random signals). The user has to define the max, min, sampling time, and seed; done programmatically.

  2. Constant inputs (defined in the workspace)

  3. Coverage method (tested for Piecewise Constant Inputs)

Question: Which tool should perform the simulations?

  1. Using Breach requires (i) identifying how the results will be saved and mapped back the simulation traces to ref, u, y so that the Neural Net training is done,
  2. Use Breach to define the reference traces and store the values (reference traces for most examples). Then, load (use from workspace block) to Simulink for each scenario and simulation trace.
    Probably, option 1 is easier but it involves linking the tools beforehand and fixing all naming, variables, connectivity issues.

To DO

  • Adding a new inport block in Simulink model files

  • Select the input stimulus, e.g. fixed constant, varying time steps, various interpolation methods, etc.

  • Do the linking among the tools.

  • Are the variables names correctly found by Breach? How to manually select the variables for Breach?

  • Write a new function which calls Breach, selects the variable names from the Simulink diagram, and most importantly chooses the input stimulus.

@nikos-kekatos nikos-kekatos self-assigned this May 3, 2020
@nikos-kekatos nikos-kekatos added the enhancement New feature or request label May 3, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant