Adaptation of an auto-generated code using a model-based approach to verify functional safety in real scenarios
Abstract
The level of autonomy of our vehicles is rapidly increasing. However, the acceptance of fully Autonomous Vehicles (AVs) depends on the confidence in their ability to operate safely in an uncontrolled environment. Hence, experts and nonexperts must have a rigorous method along with adequate tools that can support their exigencies and safety specifications. This paper presents a Domain-Specific Modeling Language (DSML) for defining formal rules and generating flawless artefacts, which enables the application of a Safety Analysis of Violations and Inconsistencies (SAVI). The validity of the approach is illustrated on a Renault use case implementation with formal safety goals for autonomous vehicles. Our approach allows designers to detect violation ambiguities and rule inconsistencies on real or simulated scenarios. Index Terms-Autonomous vehicles, safety rules, model-based system engineering, formal methods, requirement engineering, model development and verification, test and simulation.
Origin : Files produced by the author(s)