AI & Big Data
Data volumes in vehicles have been growing for years, and automated driving will increase them further. This is why IAV uses big data methods and artificial intelligence in development work, for fleet analysis, for new driving functions and in after-sales. We have access to a network of internal experts and renowned cooperation partners.
Complex data analyses not only play a key part during the development process but also in operating vehicles and technical systems – experts speak of “big data”. As data volumes increase, however, it becomes more and more difficult to extract the really relevant information. For this reason, anyone wanting to respond quickly in everyday work must process data in a meaningful way and make it easily accessible to people. In the meantime, identifying such correlations in data has become a science in its own right. IAV has experts, including engineers, mathematicians, computer scientists and physicists, who offer our customers tailored advice and assess their data in terms of quality and potential added value.
Among other areas, we have developed extensive expertise the field of intelligent control unit algorithms, making us one of the technology leaders in the domain of predictive analytics/predictive maintenance in the automotive sector. We use big data techniques for identifying patterns in sensor data with a view to deriving a probable future behavior. We also offer this expertise to customers from other sectors.
IAV’s approach to dealing with increasing data growth consists in an end-to-end big data value chain that goes over and beyond the conventional recording and analysis of measurement data. IAV’s high-performance measurement data platform and powerful analysis framework not only offer real-time processing of huge amounts of data but also supply intelligent decision aids for operative and strategic projects. IAV’s measurement data platform with interactive dashboards provides a detailed status of your vehicles at any time. Integrated in a web application, the numerous features such as measurement equipment and configuration management, data analysis or live data view form an excellent management system. The range of functions also includes the development of various algorithms for pattern recognition in time series data as well as applications in the fields of data mining and machine learning. Our high-performance analysis framework helps to process and analyze different data formats in next to no time. Other analysis tools can also be integrated depending on customer requirements.
In addition to big data, the use of artificial intelligence (AI) is one of the major trends in engineering. AI can be used in very different areas: in development for optimizing processes; in vehicles to permit new functions such as automated driving or predictive maintenance; and in the mobility sector for optimizing traffic flows and interconnect various providers in the best possible way.
In recent years, IAV has been working on all of these aspects, building up an excellent network of experts both in-house and with partners. Our own AI community has an AI methodology toolkit that can be used throughout the company and expands its knowledge at regular company-wide events, like BarCamps. In addition, we cooperate with renowned partners, such as the German Research Center for Artificial Intelligence (DFKI), MIT and the “Cyber Valley” research partnership. Together with DFKI, we operate the joint “Learning from Test Data Research Laboratory” which investigates the use of deep learning in engine development. In the “Cyber Valley” research cooperation, we work with the Max Planck Institute for Intelligent Systems on self-learning processes for automotive engineering.
In the meantime, various successfully completed projects demonstrate our leading role in the use of AI. In cooperation with Renault, IAV has developed the architecture and system for automated driving at level 4 in the SYMBIOZ demonstration vehicle. IAV’s expertise provided the basis for producing scenarios, such as automatic parking and automatically passing through toll stations, as well as solutions for problematical situations, such as sensor failures or difficulties encountered by the driver in regaining control of the car. For the HEAT (“Hamburg Electric Autonomous Transportation”) project, IAV supplied the automated shuttles which will travel on public roads in moving traffic at up to 50 kilometers per hour.
Our AI expertise has already been channeled into several of our development tools. In this way, IAV Maskin guarantees that video data from real-world road traffic is pseudo-anonymized for compliance with the General Data Protection Regulation (GDPR) without any loss of information. First, the tool uses artificial intelligence to recognize key facial features, such as facial expressions and line of vision in the original data. Faces originally recorded are then replaced with synthetic counterparts with the same features.
By combining domain and AI knowledge, we can offer our customers a wide range of services and solutions from consulting and our development tools to embedded code for control units. Our line-up also includes high-performance computing (HPC). Our CPU 24/7 subsidiary develops and operates tailored on-demand services for HPC that specialize in CAE applications and are based on the latest industry standards for HPC hardware, network components and software technologies.