IV&V Facility Research Program Results and SARP Results  

Advanced Search
Domain-Specific Analysis < Verification & Validation < Results Home

Click here to complete a short survey. The results of this survey will be used to help us improve the research program and this website.

Click here to view research projects that had new research results added in the last 90 days.
Constraint Determination in Matlab_Simulink Models and Autogenerated Code
Point of Contact Michael Beims
Michael.A.Beims@ivv.nasa.gov
Dates October 2004 - October 2005
Problem The problem of determining variable constraints is not just a theoretically interesting problem but rather is a significant pragmatic problem within programming, testing, and safety assurance. Variables whose values can exceed safety and reliability limits can result in mission failure. Conversely, variables who values cannot reach portions of their specified range can prevent software, and the hardware the software controls, from providing required functionality. This research and development effort focuses on determining variable constraints with Matlab/Simulink models and the corresponding automatically generated source code produced by these models. The results of this effort will place software tools in the hands of practitioners that will quickly determine the constraints of both internal and external variables of both Matlab/Simulink models and the source code produced from these models.
Objective Constraint determination will be done leveraging existing tools for static analysis and previous research by Dr. Supratik on static analysis. The methodology is based on converting Matlab models into clauses in constraint logic programs. The methodology has already been applied to analysis of C programs. The tools allow the project to start quickly and move to testing and Beta release, and then full release, within a single year.

The project will progress as follows:
  1. Determine signal dependencies within a Matlab/Simulink model (using MATT tool)
  2. Apply constraint determination (leverage Dr. Supratik's method)
  3. Produce signal constraint values (produce file readable by Dr. Supratik's tools)
  4. Map signals to code variables (leverage Matlab/Simulink functionality)
  5. Derive code constraints (using Dr. Supratik's tools)
  6. Compare to code constraint values to model constraint values (using MATT tool)
Each step will be rigorously tested before moving the next step in the process. Verification and validation of each step will include hand verification of small models and hand verification of a sampling within larger models. The Principle Investigators are highly experienced in verification and validation of software and software tools.
Results Facility Initiative Technical Presentation.ppt
Signal constraint determination within MatlabSimulink models.doc
Signal dependency determination with MatlabSimulink models.zip
Keywords Model Checking, MATLAB, Simulink. Auto generated code.
Categories Design Analysis
Domain-Specific Analysis
Static Analysis


Curator: Josh Stonestreet
NASA Official: Lisa Montgomery
 NASA Insignia NASA Privacy, Security, Notices