Autonomous Vehicles for Urban Mobility: Progress and Open Questions
Autonomous Vehicles, Self-Driving cars, Driverless Cars have been recurring keywords in the media, and an important focus for the research community, for more than a decade now. However, as we move past the peak of the hype cycle, the industry is bracing for a development timeline that is much longer than many early predictions.
In this talk, I would like to focus on what I consider the main fundamental issues that remain essentially unresolved, and will require a concerted effort by industry, academia, and regulatory bodies to address.
These issues essentially go beyond the (very hard, but in a sense "standard" and well studied) problems of control, perception, etc. and revolve around making sound decisions on precisely how we want these vehicles to behave, both at the individual, single-car level, and at the fleet level. In other words, how we want these vehicles to behave when interacting with pedestrians, cyclists, or other cars, and what effect we want them to have on urban mobility, including, e.g., their impact on the urban environment, public transit, and society.
Far from providing a definitive answer, I will discuss some of the approaches we have been pursuing in addressing these problems, and some of the progress we have made, both at Aptiv (and previously at nuTonomy) and at ETH Zurich.
Contact: Benedikt Barthel firstname.lastname@example.org