Masterand (M/w/d) Self-learning Driving Functions - Weissach, Deutschland - Porsche AG

Porsche AG

Geprüftes Unternehmen

Weissach, Deutschland

vor 1 Woche

Geschrieben von:

Lena Wagner

beBee Recruiter

Beschreibung

Aufgaben:

Motivation: When using automated driving functions, humans continuously provide feedback by interacting with the vehicle and intervening if they are unsatisfied with the function's behavior. This feedback is currently unused but could be utilized to adapt the function's behavior to the driver's individual wishes. Such human machine cooperation can be used to further increase human acceptance of driver assistance systems and automated driving functions.
Objective: This master thesis focuses on the development of a concept for an iteratively learning driving function using driver feedback. As proof of concept, the developed algorithm should be implemented in a simulation environment, using a trajectory planner and tracker which follows predefined routes. Driver interventions, e.g. gas or braking signals, should then be used as feedback to adjust the planner's behavior in order to follow individual driver wishes. The exact structure of the thesis will be defined with the thesis student and the supervisor.
Tasks:
Research on state of the art in the area of selflearning algorithms (e.g. reinforcement learning), human machine cooperation, and trajectory planning
Concept development and evaluation of the found state of the art approaches for the use case of automated driving functions and human feedback
Implementation of the designed algorithm in a simulation environment
Test and evaluation of the developed system using prerecorded driver interventions of automated driving functions during real world drives
Duration: 6 Months

Anforderungen:

Studies in engineering or natural sciences (e.g. computer science, mathematics, physics, electrical engineering, etc.), ideally with a focus on machine learning, data science, robotics, automated/autonomous driving
Qualified knowledge in the field of machine learning Experience in path/trajectory planning
Experience with Carla Simulation advantageous
Ideally Experience with ROS (Robot Operating System)?Excellent English skills (German is a plus)
Analytical thinking, structured and independent way of working, creativty, and motivation to work on a complex task
Very good programming skills in Python Experience with at least one deep learning framework, e.g. TensorFlow or PyTorch

Unternehmensprofil:

"Am Anfang schaute ich mich um, konnte aber den Wagen, von dem ich träumte, nicht finden. Also beschloss ich, ihn mir selbst zu bauen."

Dieser Satz von Ferry Porsche bringt alles auf den Punkt, was Porsche ausmacht. Als Marke, als Unternehmen, als Automobilhersteller, als Arbeitgeber. Die "Idee Porsche" hat so einzigartige Sportwagen wie den Porsche 356 oder den 911 hervorgebracht.

Durch die Kombination von Tradition und Innovation mit den Porsche typischen, manchmal etwas unkonventionellen Denkweisen unserer Mitarbeiter, lassen wir automobile Träume wahr werden und haben so immer das "Etwas-mehr" im Blick.

Porsche ist allerdings viel mehr als "nur" ein exklusiver Sportwagenhersteller.

Denn auch als Arbeitgeber hat Porsche viel zu bieten:
vielfältige Einstiegs
- und Karrieremöglichkeiten, Maßnahmen zur Vereinbarkeit von Beruf und Familie sowie flexible Arbeitszeitmodelle.