“IAV Unites the IT and Automotive Worlds”

Whether electric mobility, autonomous driving or the trend towards “automotive devices” – the development of modern vehicles is highly complex and is changing rapidly. As a software partner, IAV offers vehicle manufacturers comprehensive support. Targeted data analyses can be used to detect errors in vehicle development at an early stage or to test new innovations. Jens Schulze, Head of Data Analytics and Fleet Validation, explains how data analyses work and why IAV is one of the market leaders in this field.

iav vereint die IT

In recent years, the importance of big data analytics in vehicle development has grown considerably. What exactly does IAV do in this field?

We provide support in vehicle development, particularly in the release of vehicle functions and the entire vehicle. This is an area in which we have been active for over 20 years and for which we have developed a special service – the IAV Measurement Data Platform. At its core, it deals with the storage, analysis and visualization of vehicle data.

What exactly is this data and at what stage of vehicle development does this work take place?

Our work takes place in the field of fleet validation. This means that we work on prototypes, pre-series or end-customer vehicles that already have most of the planned vehicle functions and are tested again for functionality before they are released. First, we collect the raw data from the vehicle and store it in a specially equipped data center. There, we have the necessary technical prerequisites to keep the data as efficient as possible and to analyze it with regard to certain functionalities.

«Our analyses are high performance, efficient and customizable. We thus offer an all-round package.»

Jens Schulze — Head of the Data Analytics and Fleet Validation Department at IAV

What makes IAV stand out when it comes to analyzing vehicle data and what added value does it offer customers?

Data analyses in the automotive sector require very detailed vehicle and IT knowledge. This is because the interconnectivity of vehicle electronics has become increasingly complex in recent years. Communication between the electronic participants is handled via so-called services, which require a great deal of computing power when the system is heavily utilized. If this computing power is not available, errors in the information flow can occur. Finding these errors is not so easy due to the enormous amount and complexity of data. However, since we have a very precise knowledge of automotive engineering and the IT world, we are one of the market leaders in this field. Our analyses are high performance, efficient and customizable. This enables us to offer the customer an all-round package.

You use a special IT architecture for this. How exactly is this structured?

Before we start a project, we first record the customer’s requirements and objectives as well as all processes and use cases. In the meantime, we have developed a very extensive range of algorithms and models that cover almost all analyses in the automotive sector. Then we decide which components are needed and build the appropriate IT architecture according to the requirements. There is a choice of various infrastructure and connectivity concepts, virtualization approaches and application layers as well as analytics framework. In the case of infrastructure and connectivity, for example, the question is whether a private, public or hybrid cloud is used. For virtualization, we choose between container technologies, virtual machines or serverless models. The analytics framework are based either on GPU or CPU processors and the architecture is based on a streaming, batch or lambda approach.

IAV vereint die IT- mit der Automotive-Welt

What analyses can be used by customers within the framework of the IAV Measurement Data Platform?

The platform’s IT architecture and analysis models were developed specifically for use in the automotive sector and are technically state-of-the-art. They enable us to map everything that is currently possible. This includes above all vital and real-time data from the vehicle, but also the mapping of specific forecasts and use cases from endurance testing. The results of these analyses can be visualized in different ways. We have a very comprehensive, modular system that allows customers to configure their own dashboard. The dashboard gives them a simple overview of all errors and anomalies and helps them to improve vehicle functions and ECUs.

Can you give a specific example from your everyday work?

A classic case would be pattern recognition in case of a communication error of the ECUs. Let’s say that after all software components have been integrated, the brake light no longer works under certain conditions. In daily operation, it would be very costly for a manufacturer to detect this error themself. But with our analysis models it is no problem, because they know the software architecture of the vehicle very well and the customer can access it via a web-based measurement data management system. Thus, the error can be found within a very short time via a simple search function and displayed in the desired form. Very similar to a Google search.

«So our unique selling point is that we combine the IT and automotive worlds. We are experts in our field […]»

Jens Schulze — Head of the Data Analytics and Fleet Validation Department at IAV

The article was published in automotion 03/2020, the automotive engineering magazine of IAV. Here you can order the automotion free of charge.

Stay up to date

Subscribe to the newsletter