Ghada Bahig

Abstract

Software applications are increasingly dominating safety critical domains. Safety critical domains are domains where the failure of any application could impact human lives. Software application safety has been overlooked for quite some time but more focus and attention is currently directed to this area due to the exponential growth of software embedded applications. Software systems have continuously faced challenges in managing complexity associated with functional growth, flexibility of systems so that they can be easily modified, scalability of solutions across several product lines, quality and reliability of systems, and finally the ability to detect defects early in design phases. AUTOSAR was established to develop open standards to address these challenges. ISO-26262, automotive functional safety standard, aims to ensure functional safety of automotive systems by providing requirements and processes to govern software lifecycle to ensure safety. Each functional system needs to be classified in terms of safety goals, risks and Automotive Safety Integrity Level (ASIL: A, B, C and D) with ASIL D denoting the most stringent safety level. As risk of the system increases, ASIL level increases and the standard mandates more stringent methods to ensure safety. ISO-26262 mandates that ASILs C and D classified systems utilize walkthrough, semi-formal verification, inspection, control flow analysis, data flow analysis, static code analysis and semantic code analysis techniques to verify software unit design and implementation. Ensuring software specification compliance via formal methods has remained an academic endeavor for quite some time. Several factors discourage formal methods adoption in the industry. One major factor is the complexity of using formal methods. Software specification compliance in automotive remains in the bulk heavily dependent on traceability matrix, human based reviews, and testing activities conducted on either actual production software level or simulation level. ISO26262 automotive safety standard recommends, although not strongly, using formal notations in automotive systems that exhibit high risk in case of failure yet the industry still heavily relies on semi-formal notations such as UML. The use of semi-formal notations makes specification compliance still heavily dependent on manual processes and testing efforts. In this research, we propose a framework where UML finite state machines are compiled into formal notations, specification requirements are mapped into formal model theorems and SAT/SMT solvers are utilized to validate implementation compliance to specification. The framework will allow semi-formal verification of AUTOSAR UML designs via an automated formal framework backbone. This semi-formal verification framework will allow automotive software to comply with ISO-26262 ASIL C and D unit design and implementation formal verification guideline. Semi-formal UML finite state machines are automatically compiled into formal notations based on Symbolic Analysis Laboratory formal notation. Requirements are captured in the UML design and compiled automatically into theorems. Model Checkers are run against the compiled formal model and theorems to detect counterexamples that violate the requirements in the UML model. Semi-formal verification of the design allows us to uncover issues that were previously detected in testing and production stages. The methodology is applied on several automotive systems to show how the framework automates the verification of UML based designs, the de-facto standard for automotive systems design, based on an implicit formal methodology while hiding the cons that discouraged the industry from using it. Additionally, the framework automates ISO-26262 system design verification guideline which would otherwise be verified via human error prone approaches.

Department

Computer Science & Engineering Department

Graduation Date

2-1-2018

Submission Date

July 2017

First Advisor

El-Kadi, Amr

Committee Member 1

Salem, Ashraf

Committee Member 2

Hammad, Sherif

Extent

265 p.

Document Type

Doctoral Dissertation

Rights

The author retains all rights with regard to copyright. The author certifies that written permission from the owner(s) of third-party copyrighted matter included in the thesis, dissertation, paper, or record of study has been obtained. The author further certifies that IRB approval has been obtained for this thesis, or that IRB approval is not necessary for this thesis. Insofar as this thesis, dissertation, paper, or record of study is an educational record as defined in the Family Educational Rights and Privacy Act (FERPA) (20 USC 1232g), the author has granted consent to disclosure of it to anyone who requests a copy.

Institutional Review Board (IRB) Approval

Approval has been obtained for this item

Download

Share

COinS