Presented By O'Reilly and Cloudera
Make Data Work
22–23 May 2017: Training
23–25 May 2017: Tutorials & Conference
London, UK

How to prevent future accidents in autonomous driving

Dr.-Ing. Michael Nolting (Volkswagen Commercial Vehicles)
17:2518:05 Wednesday, 24 May 2017
Secondary topics:  AI, IoT, Logistics, Streaming
Level: Beginner
Average rating: *....
(1.67, 6 ratings)

Who is this presentation for?

  • Data scientists, data engineers, big data architects, and automotive fans

What you'll learn

  • Learn a generalized method for anomaly detection and a real-time architecture for autonomous driving based on the Kappa architecture (Flink, Kafka, Elasticsearch, and Kibana)
  • Gain insights into automotive data and the automotive industry

Description

Tesla calculated that its fleet has to drive 10 billion kilometers to train and calibrate its machine-learning algorithms to drive autonomously. The challenge for autonomous driving, however, is that the algorithms can hardly be initially trained with all possible driving conditions—due to the highly dynamic runtime environment for the sensors—and dangerous situations only occur at a very low frequency. This problem cannot be solved by increasing the amount of training data, since there will be driving situations where no data will be available due to disturbed or noisy sensor output. It requires a monitoring system able to detect regions where the probability is high that sensors will fail.

To overcome this problem, a system has to be created where cars, regardless of their brand, sample their environment during operation and dangerous situations are detected by a central server and propagated to other cars in real time, a process much like website monitoring. Michael Nolting explores how such a system might be realized (with a real-time architecture based on Kafka, Flink, Elasticsearch, and Kibana) and how its requirements can be fulfilled. Michael shares a generalized method for detecting anomalies (dangerous situations) based on bootstrapping, which is able to operate on non-Gaussian distributions. Bootstrapping allows assigning measures of accuracy (defined in terms of bias, variance, confidence intervals, prediction error, or some other such measure) to sample estimates and enables estimation of the sampling distribution of almost any statistic using random sampling methods

Photo of Dr.-Ing. Michael Nolting

Dr.-Ing. Michael Nolting

Volkswagen Commercial Vehicles

Michael Nolting is a data scientist for Volkswagen commercial vehicles. Michael has worked in a variety of research fields at Volkswagen AG, including adapting big data technologies and machine learning algorithms to the automotive context. Previously, he was head of a big data analytics team at Sevenval Technologies. Michael holds a Dipl.-Ing. degree in electrical engineering and an MSc degree in computer science, both from the Technical University of Brunswick in Germany, and a PhD in computer science.