Deep Reinforcement Learning Control for Autonomous Vehicle Driving

  • Posted on: 19 February 2021
  • By: lottakaka
  • Updated on: 22 February 2021
Type recrutement: 
Master
Section: 
27 (Informatique)
61 (Génie informatique, automatique et traitement du signal)
Entité et lieu: 
R&D Team in Siemens Digital Industries Software, Leuven, Belgium
Détails: 
In this project, we investigate deep reinforcement learning for autonomous vehicle control. The learning process is based on digital twin models of traffic environment, vehicle dynamics, and physic-based sensor (lidar, camera).
 
We are looking for outstanding students who are eager to do their Master thesis or Internship on a reinforcement learning (RL) topic for autonomous driving applications in a dynamic and international research environment. Examples of possible (but not limited to) topics:
- Traffic agents modelling and motion prediction
- Deep RL control, curriculum learning
- Transfer learning from simulation to physical vehicle control
 
The company provides various tools to support the research activities, for example, Siemens Prescan for sensor and traffic modelling, Amesim for vehicle dynamics modelling, autonomous vehicle setup for embedded control, and other autonomous driving platforms for deep learning, sensor fusion implementations.
 
Candidate profile: Background in machine learning, computer vision, robotics/control, and familiar with programming. Experience with ROS, OpenAI or autonomous vehicles is a plus.
 
 
Contact email: son.tong@siemens.com