Automated driving, electric mobility, connectivity and digitalization are our core topics.
Our research and development (R&D) activities focus on developing innovative and sustainable products, systems and services for our customers in a wide variety of industries.
As part of the preparations for the new organizational structure, which will be implemented from 2020 onwards, we have been working on the design of the new central Automotive R&D function since the beginning of 2019. This new area will incorporate the development functions of our present Interior and Chassis & Safety divisions as well as those of our current central functions. By the end of the year, autonomous driving and connected mobility technologies will be combined under the roof of Automotive R&D. Our software and hardware engineers will form a global center for predevelopment and application development.
The new area will strengthen our cross-organizational collaboration, shorten innovation cycles and further enhance the flexibility of our innovation processes – particularly in relation to software development. Our customers and end users will benefit from state-of-the-art, affordable solutions that help to prevent accidents, bypass traffic jams and increase driving comfort.
The Powertrain division’s R&D locations have stayed virtually the same. Their areas of focus include combustion engine, hybrid and all-electric drive systems – including battery activities. The
R&D organizations of the Tire and ContiTech divisions will remain unchanged by the future organizational structure. R&D activities in the ContiTech division have a largely decentralized structure by virtue of the different product segments. The central Innovation & Digitalization unit and the central Business Development unit that the ContiTech division set up over the course of the reporting year have the goal of fostering innovative products and enhancing the existing portfolio with new services including mobility services. Product requirements for tires are very similar worldwide, which is why R&D has a mostly centralized structure. For example, our R&D site in Hanover-Stöcken has around 1,400 employees working on the development of up to 9,000 different tires to meet various requirements with regard to speed rating approvals, rolling resistance optimization, inch dimensions and application purpose. Our international scouting system ensures that we pay sufficient attention to the requirements of local markets.
|Research and development expenses (net)|
|€ millions||% of sales||€ millions||% of sales|
|Chassis & Safety||1,023.2||10.7||913.8||9.4|
|Capitalization of research and development expenses||158.0||92.1|
|in % of research and development expenses||4.7||2.9|
|Depreciation on research and development expenses||90.0||74.5|
Machine-learning advanced driver assistance system
We completed an exceptionally complex project during the reporting year: PRORETA 4, a three-and-a-half-year research project carried out in partnership with the Technical University of Darmstadt. The project’s aim was to develop a machine-learning vehicle system (City Assist System) to help drivers navigate inner-city traffic. The system is already being used as a prototype. Radar sensor data helps the system to assess the traffic situation when making a left turn, entering a roundabout or approaching a right-before-left intersection. Machine-learning technology played an instrumental part in the project.
To enable an assistance system in a complex driving situation to give the driver a recommendation that the driver will accept – and to become familiar with the driver as a good passenger would – the system needs to analyze the driver’s driving style and subjective perception of safety and risk. Using a machine learning method is a quick and reliable way of developing this kind of driving profile in which the system analyzes data that is recorded during the process of driving. Acceleration, direction of movement, braking maneuvers and lateral acceleration are all things that provide the algorithm with information on the type of driver it is dealing with.
Extensive test drives with test subjects have revealed that the algorithms used in the City Assist System are able to make conclusions about the driver’s current driving style after three to five driving maneuvers. The driver is then assigned to one or several clusters of driving profiles, which allows the City Assist System to personalize its driving recommendations.
Continental’s global research network for artificial intelligence (AI) continued to expand in the year under review. After the University of Oxford, DFKI (German Research Center for Artificial Intelligence) and other organizations, Continental signed an agreement with the AI research group Berkeley DeepDrive (BDD) at the University of California. This partnership focuses on optimizing the speed of neural networks in cars, as well as protecting AI systems in safety-critical applications. The AI research results should make their way into production as quickly as possible.
Research and testing laboratory for dandelion rubber opened
During the reporting year, we opened the Anklam Taraxagum Lab – a research and development laboratory in Anklam, Germany. The lab will continue its research into the cultivation and processing of the Russian dandelion plant as an alternative raw material source to rubber harvested from rubber trees. The plan is to be using dandelion rubber in volume production and generating a growing percentage of our natural rubber supply from dandelion plants within a ten-year timeframe. We see the Russian dandelion plant as an important alternative and supplement to conventional natural rubber as it will enable us not only to meet the growing global demand for rubber by reliable means, but also to make tire production more sustainable and environmentally friendly.
First self-driving tire-testing vehicle
Our first self-driving vehicle for testing tires on a variety of surfaces is now operational at our test track in Uvalde, Texas, U.S.A. Our aim is to further enhance the validity of test results for Continental passenger and light truck tires and minimize the impact of the test process on the results themselves. The new test vehicle is controlled with the help of a satellite-based positioning system and is based on Continental’s Cruising Chauffeur, which was developed for automated driving on freeways. Automated vehicles allow us to reproduce processes accurately so that every tire undergoing testing is subjected to exactly the same conditions. This means that we can reliably determine that any differences in the test results are actually due to the tires themselves and not to the test procedure.
Intelligent solutions for conveyor belts
To demonstrate the different conveyor belt service options that exist for bulk materials and piece goods, we have developed a model that illustrates the latest market trends for belt monitoring as well as full-service applications. Our solutions are equipped with sensors that monitor every movement the conveyor belt and the conveyed material make. They inspect surfaces, report load levels and identify mistracking belts in real time. The information is stored in databases and analyzed by algorithms, which know when the belt needs servicing. Furthermore, there are monitoring systems in place to inspect the belts’ safety-related properties.
The technology also meets the conditions required for new business models like “pay per ton” and anticipatory maintenance of components and systems.
We are already in a position where our customers are able not only to purchase a belt, but also to put together an end-to-end package comprising conveyor belts and services.