摘要: |
Michigan Department of Transportation (MDOT)’s goals center around safety and mobility on our road network for vehicles, pedestrians, and non-motorized users. MDOT’s operational practice when identifying/meeting potential challenges at Michigan state trunk line intersections has been reactionary in nature. Any planned and/or response activity resulting from crash and fatality data is delayed up to a year or so waiting for the needed data.
An emerging technology, “Video Analytics,” may provide the ability to operate in a proactive way by identifying crash challenges in real time and to help form an understanding of near misses and crashes by utilizing a video software analytics platform. The software can predict crashes for all users ahead of time and can identify challenges that need expert review to develop solutions to save lives.
If MDOT could obtain real time information about near misses happening at intersections for review by a system expert, then change could be made before the near misses turn into crashes. The software system has an analytics platform that utilizes real time system queries, including a multitude of sorting capabilities. Additionally, the analytics system can look down a given corridor to pick up potential impacts mid-block.
This research would install the video analytics system along a test corridor to evaluate, analyze and validate the effectiveness and improved traffic efficiencies in implementing solutions at MDOT’s most challenging signalized intersections. The system could take this information and provide a notification through connected vehicle technology to vehicles, pedestrians, and non-motorized users. Initially, this notification capability would be set up in test vehicles. If proven effective, MDOT could expand installations to other vehicles time. |