Workshop on Scene and Situation Understanding for Autonomous Driving (0.5d)

Organizers: Igor Gilitschenski, Juan Nieto, Federico Tombari, Daniela Rus


Enabling robust higher level scene and situation understanding is one of the key challenges to unlock the full potential of autonomous driving. Most autonomous driving research has considered the scientific problems involved in this challenge as a special instance of either the perception or the planning tasks. This workshop takes a scene and situation centric approach to discussing advances and future directions of autonomous driving research. Our goal is to bridge the gap between perception, planning, and control-based approaches to scene and situation modeling. On the one hand, we want to discuss how higher-level scenery information can be used to improve the entire autonomy stack involving, localization, detection, planning, and control systems. On the other hand, we are interested in the interplay of classical perception, planning, and control approaches for obtaining an improved scenario understanding. In that context, we also discuss the impact on how recent advances in Deep and Reinforcement Learning can be leveraged for impacting basic research and actual deployment of autonomous vehicles.