ABOUT NORCOM
Infrastructure for driver assistance systems
The task
Driver assistance systems electronically support the driver of a car in certain driving situations. The aim is to increase the security and reliability of the systems.
The challenge
The development of such systems requires a novel architecture for storing, processing and analyzing large amounts of data. Data must be made available for processing quickly - even if it is generated worldwide or stored locally. Likewise, large amounts of data (petabyte range) must be able to be processed quickly - by different teams, regardless of location.
our solution
The central solution component is the creation of automated, robust and scalable workflows for the conversion and provision of data and for the automated evaluation of abnormalities as well as the connection to advanced analytics and machine learning for further processing in self-service. The basis is provided by the CI / CD routes developed in the project, with which software products are automatically and robustly rolled out in a standardized manner across several locations around the world. Existing analysis tools were integrated via middleware layers and thus raised to a big data-capable level.
The customer benefit
Large series of measurements generated worldwide are now immediately accessible for processing. Anomalies in processes and data arerecognized early and operational failures are avoided. Flexible and customizable analysis solutions ensure the required high speed of innovation.
Project-
Characteristics
Our role
Support of the customer by architects, data scientists, data engineers and software developers
Our activities
Parallelization of existing analysis tools
Persistence of the analysis results
Export of relevant measurement signals and provision for existing visualization tool
Technologies & methods
Applications: Jupyter, Pycharm, Zeppelin, Jenkins, Bitbucket, Artifactory, Jira, Confluence, Sonar, DaSense, Grafana
Data / databases: Hbase, Hive, OpenTSDB, MDF, ADTF, Parquet
Languages / frameworks: Python, Java, Airflow, Hadoop / MapR, Spark, Yarn, Oozie, Docker, Tensorflow, Kerberos, R-Studio
Methods: time series / image analysis, deep learning, CI / CD