Speaker Details

Speaker Company

Christian Contreras Campana

Christian Contreras Campana studied Physics and Mathematics at University of California Los Angeles (UCLA), and Rutgers University (RU), USA. He received a PhD in High Energy Experimental Particle Physics in 2015. He is currently working at Automotive Artificial Intelligence (AAI), Berlin, Germany, as an Expert Research Scientist as part of the Intelligent Traffic Simulation team. Research interests include scene understanding, behavioral planning for autonomous driving, as well as imitation learning and reinforcement learning.


Traffic simulation - modeling high-level driving decisions from naturalistic driving-data

Traditional driving stacks have inherent shortcomings in generating trajectories that are both feasible and naturalistic at high and low abstraction levels. In this presentation we propose a new model for simulated traffic agents that is able to mimic high-level decisions from naturalistic traffic data and can employ any trajectory planning method. The agent model focuses on human-like behavior, decision-making, and driving performance based on a supervised machine learning approach. We will show that even though our approach is based on modeling nanoscopic traffic-agent characteristics, realistic traffic behavior can be achieved both at the microscopic and macroscopic levels.