EAGER: Real-Time: Intelligent Intersections

点击翻译按钮获取中文摘要

Intersections are dangerous. Threats arise from interactions among pedestrians, bicycles and vehicles; traffic signal operation that does not reveal who has the right of way; blind spots from vehicle obstructions; and illegal movements. These threats cannot be avoided by autonomous vehicles, human drivers, cyclists and pedestrians who do not have the information they need to avoid wrong decisions, leading to crashes. This missing information can be calculated and communicated by an intelligent intersection. The information gives the full signal phase, an estimate of when the phase will change, blind spot occupancy, and detection of red-light violators. The proposed research will develop the real-time learning and decision algorithms of the intelligent intersection.The euphoria over the rise of the autonomous vehicle industry has concealed the fact that the safety record of autonomous vehicles today is much worse than that of human drivers. Nor is it acknowledged that the safety of autonomous vehicles and others (drivers, cyclists and pedestrians) can be improved by information generated from the infrastructure. The proposed research aims to counter these misconceptions by showing how intersection safety can be enhanced by infrastructure assistance. The research is motivated by current concerns and grounded on assumptions that are valid today. The U.S. registered 40,327 traffic deaths in 2016, an increase of 13.5% since 2010. Our aim is to make a dent in this national health emergency by helping to avoid errors that can be prevented with better intelligence. California's DMV has begun to issue permits to test vehicles without a safety driver. DMV regulations give local cities some latitude in determining the conditions under which such tests are permitted. The safety of their citizens will be of foremost concern to local authorities. This research will help them in their decisions and suggest the kind of information that they may require from tests conducted in their cities.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.