AUTOSAR gears up for autonomous driving

The main game changers for the development of automotive electric/electronic (E/E) architectures will be future develop­ments and new applications in the fields of connectivity as well as highly automated and autonomous driving.

In response, the new AUTOSAR Adaptive Platform is being specially designed to meet resulting key requirements, such as high data processing capacities and update “over the air” capability.

Automotive Industries (AI) asked Stefan Rathgeber, Head of AUTOSAR Center, Corporate Systems and Technology at Continental Automotive, and AUTOSAR Spokesperson, what is the progress on the AUTOSAR Adaptive Platform?

Rathgeber: I am confident that the initial release 1.0.0 will take place as intended at the end of the first quarter of 2017. This early release will offer a set of minimum required features. Based on market requirements and feedback, enrichment of features will follow with short release cycles. We have established the basic architecture and are currently finalizing the work on the core interface layers for applications, the middleware and the operating systems AI: Are any of the features available already?

Rathgeber: AUTOSAR is aiming to reuse and extend existing technologies in the automotive and related industries depending on the feature discussion within the working groups. Internally we are already working on a first baseline containing prototype solutions and feasibility studies. But publications are not planned before our first release date by Q1/2017.

AI: How will it be integrated with the AUTOSAR Classic Platform?

Rathgeber: With the Adaptive Platform AUTOSAR aims to provide an optimal standardized software framework for new applications in the fields of connectivity and highly automated and autonomous driving. But one of the main drivers for AUTOSAR is to keep the existing and the new technologies compatible. So AUTOSAR will care that classic and adaptive applications can be seamlessly combined at least on Ethernet level. AI: Are you continuing to develop the Classic Platform?

Rathgeber: The Classic Platform remains the system of choice with regard to control units from the classical e/e domains of the automobile. It continues to be the preferred solution for applications that make limited demands on hardware resources while fulfilling safety requirements and providing hard real-time capabilities. Upcoming new functionalities will be realized depending on the market need by the classic platform or by the Adaptive Platform, which is designed to meet the special requirements for dependable high performance ECUs.

AI: What impact will the new platforms and systems have on hardware suppliers?

Rathgeber: The Adaptive Platform will be a dynamic system that includes the middleware, a support for complex OS via a POSIX interface and state-of-the-art multicore microprocessors. This is a huge benefit for our users as multicore systems will become more and more common in the automotive domain in the future. The Adaptive Platform is designed as a hardware-independent architecture and the main communication approach is based on service oriented communication and IP/Ethernet. Thus, the platform will be capable of supporting adaptive software deployment and interaction with non-AUTOSAR systems.

AI: Are hardware suppliers involved in the rollout of the Adaptive Platform?

Rathgeber: AUTOSAR is a cooperation of more than 180 companies, including automotive manufacturers and automotive suppliers as well as tool and semiconductor vendors. Among our core and premium partners are major suppliers such as Bosch, Continental, Delphi and Valeo. These partners continuously contribute their expertise in our working groups during the entire process.

AI: Where to from here?

Rathgeber: In my opinion, there will be four main drivers for new automotive software systems. Highly automated driving will require dependable systems that include fail-safe operational systems, while car-2-X applications demand the interaction of vehicles and off-board systems. With regard to vehicles in the cloud scenarios, we will need to take dedicated means for ensuring security into account. The increased interconnectivity will also lead to a stronger interaction between automotive software systems. As a result, we need to consider non-AUTOSAR and off-board systems in our methodology.