Neural Networks for Sensor Fusion in Autonomous Vehicles (Part II)

8.3387

Dozenten

Beschreibung

The goal of this study project is to evaluate and prototype applications of current deep learning techniques for sensor fusion in autonomous vehicles. Therefore, application scenarios such as road quality detection, active chassis control, wear monitoring, and smart sensor interpretation should be realized. Facilitating the use of the prototyped applications on real roads, the worst-case data input of the prototyped neural networks should be predicted in order to know situations where the networks fail.
In the project, we will focus on deep learning techniques and sensor fusion. We will analyze recent developments in the field, evaluate selected approaches and check their potential for autonomous driving. Based on this analysis, we will develop prototype systems and test it on real data sets and depending on the results in real vehicles.
The project will be conducted in cooperation with a partner from local industry, who can provide a broad set of training data and an existing system for baseline evaluation.

Prerequisite:
The ideal participant in this project has some background in machine learning and computer vision, as well as a basic knowledge of neural networks. Some experience with deep learning frameworks (TensorFlow, etc.) and a general fascination for autonomous vehicles would be of additional benefit.

Weitere Angaben

Ort: (50/219): Freitag, 26.04.2019 16:00 - 18:00, (50/216): Freitag, 03.05.2019 16:00 - 18:00
Zeiten: Do. 14:00 - 16:00 (wöchentlich), Termine am Freitag, 26.04.2019, Freitag, 03.05.2019 16:00 - 18:00
Erster Termin: Freitag, 26.04.2019 16:00 - 18:00, Ort: (50/219)
Veranstaltungsart: Studienprojekt (Offizielle Lehrveranstaltungen)

Studienbereiche

• Cognitive Science > Master-Programm