For a better experience on dSPACE.com, enable JavaScript in your browser. Thank you!

Use Case: Validating AUTOSAR Architectures

Although AUTOSAR provides a standard for exchanging system description files, difficulties often arise during data exchange. This holds true for both the data exchange between OEMs and suppliers and the data exchange between different AUTOSAR tools. Early in the development process, it is often not automatically checked whether the data is correct. This means that incomplete or inconsistent ARXML files might be exchanged. This, in turn, results in a time loss and additional work, as the errors have to be reported, fixed at their source, and the exchange has to be repeated.

In order to achieve a higher level of quality before data transmission, a manual or automatic validation of the data is needed. Ideally, this is done long before the date exchange.

Validation with SystemDesk

  • Comprehensive set of validation rules
  • Project-specific validation rules can be created
  • Integration into automatic validation processes
  • Report generation with test results
  • Complete support of the AUTOSAR schema

Examples of Data Exchange

Examples of data exchange in the AUTOSAR development process.

Between OEM and Supplier

The data exchanged between OEMs and suppliers typically consists of system description files. Before a file is transmitted, it must be verified as complete according to the particular company process. Examples for incomplete data are missing data types or init values, or missing elements in the bus specification. For this data exchange, the data is typically specified to only a certain level of detail, which varies between the companies.

In addition to completeness, semantic correctness has to be checked. This can be done by using AUTOSAR constraints, which define semantic rules based on the AUTOSAR schema. It makes sense to also include additional semantic checks, which can again be partly company-specific. For example, semantic checks can be used to verify whether the init values lie in the value range defined for the relevant data type.

Between Architecture Tool and RTE Generator

When data is exchanged between tools, different conditions apply. An RTE generator, for example, requires complete data for everything that affects the RTE. This means that an implementation data type must be defined for each application data type, a detailed specification of the data access from runnables to variables is needed, etc.

As for the interfaces between different parties, in addition to completeness semantic correctness is also a crucial factor for the data exchange between an architecture tool and an RTE generator. Here, too, the AUTOSAR constraints are a starting point for finding semantic errors and can be extended to include further checks.

Automatic and Manual Validation

Validating with SystemDesk is made easy by precise error messages.

For AUTOSAR-based development, dSPACE offers the powerful tool SystemDesk®. It provides comprehensive validation features to avoid tedious correction iterations. In general, validation should be performed as often as possible while data is being created, as this lets users detect and eliminate errors very early on.

SystemDesk has a comprehensive standard set of rules that cover a large proportion of everyday use cases. The included rules contain many constraints of the software component template and the system template according to AUTOSAR as well as completeness checks for RTE generation.

This enables users to start the validation directly without a lot of effort. It is also possible to select only some of the predefined rules to use them in a particular scenario. Additionally, users can easily add their own rules to the existing ones via Python scripts. This comes in handy for two reasons: First, this is the only way to ensure completeness in a company-specific process. Second, this makes it possible to create rules for a particular tool: e.g., for a specific RTE generator that expects specific data.

A comprehensive validation operation reliably detects and eliminates errors very early on. For smaller use scenarios, the validation can be started manually, giving users full control of each step. For larger use cases involving a high number of people and files, SystemDesk allows for an automated process that validates the files during each check-in.

In both cases, a validation report provides information about all found errors, including detailed information with precise and helpful error messages as well as direct links between the error description and the involved component.

In addition to the validation support, SystemDesk is very open when it comes to importing ARXML files. All files that were built according to the AUTOSAR schema are imported and exported in their entirety. This way, SystemDesk does not create new obstacles for importing files. Using SystemDesk for validation therefore does not create another bottleneck that the tool chain has to be adjusted to.

Further Information