27
.
11
.
2018
-
Meetup

IoT Data Solutions for Connected Cars

Diese Veranstaltung ist leider schon vorbei!

Information from connected devices such as cars can be used for many different purposes like gaining better understanding of the product use, identify and forecast potential product issues or design new generations of products. New driver assistant functions, like early warnings about accidents or icy roads as well as high traffic or free parking spots are based on anonymized realtime fleet information.


Talk 1:
Collecting Telemetry Data: Being fast - but not too fast
Speaker: Dr. Andreas Schroeder, Kai Rusteberg (BMW Group)
30 Minutes

Telemetry data can be a vital source of information for engineering and for coordinating maintenance activities. Having a high-throughput data channel is therefore an important asset for the BMW Group.

In this talk, we will present how we redesigned an existing telemetry data collection backend system to deliver the speed and throughput needed. Being fast is however not the only virtue that had to be achieved: we will also discuss why adhering to speed limits is of paramount importance, and how we designed our system to achieve this goal.

Dr. Andreas Schroeder is currently working as IT specialist in the vehicle data collection group. He has been previously working at AutoScout24 GmbH as lead developer. He has been designing high-throughput and low-latency data backends, and implemented microservices and cloud infrastructure on AWS. He also worked as IT consultant at codecentric AG, consulting companies on DevOps principles, cloud infrastructure and microservice system design. He holds a PhD in Computer Science from Ludwig-Maximilians-Universität München.

Kai Rusteberg is currently working as IT specialist in the vehicle data collection group. He has been previously working at codecentric AG as lead developer and senior IT consultant. He has been designing reactive distributed systems for highly scalable data backends on top of Cassandra and Elasticsearch.


Talk 2:
Scalable Microservice Pipelines on AWS
Speaker: Dr. Lars Haferkamp + Riccardo Valentini (Comsysto Reply)
30 Minutes

Most driver assistant functions, like distance warning, sign recognition or parking assistant are built on the immediate environmental model of the car’s sensors. New functions, like early warnings about accidents or icy roads as well as high traffic or free parking spots need an extension of this environmental model. Additionally highly automated cars will need an extended environmental model beyond their own view to be able to act in a multi-agent environment.

As new cars of the BMW Group’s fleet drive and track their environment they send their sensor data via a secure connection over the mobile network to a common backend. This enables the aggregation and extraction of new information, which can be send back to other drivers. We show how this kind of streaming data is processed and monitored in the AWS cloud by scalable microservice pipelines.

Dr. Lars Haferkamp is working as a Data Scientist and Engineer at Comsysto Reply on several AWS cloud-based sensor data collection and analytics solution with the BMW Group. Before joining Comsysto Reply he was working for Deutsche Telekom as a Software Consultant.

Riccardo Valentini is Computer Scientist with strong focus on Software Development and Agile Methodologies, who has experience with Machine Learning and Data Mining. His special interests are large scale data applications, cluster computing and Blockchain technologies.

maps:Riesstr. 25, 80992 München
Datum:
27.11.2018
Uhrzeit:
19:00 Uhr
Ort:
ORACLE Deutschland B.V. & Co. KG Hauptverwaltung Riesstr. 25, 80992 München
Speaker:
Trainer:
Tickets:

Sprecher

No items found.

Anmeldung

Die Anmeldung erfolgt über Meetup. Klicke einfach auf den unten stehenden Button, um auf die Veranstaltungsseite weitergeleitet zu werden.

Die Anmeldung erfolgt über Meetup. Klicke einfach auf den unten stehenden Button, um auf die Veranstaltungsseite weitergeleitet zu werden.

Anmeldung