Google researches new AI model that can improve traffic efficiency by 30%

Google Research recently published an article introducing the application results of the "traffic guidance" AI model they developed using the open source simulation software SUMO (Simulation of Urban Mobility)

It is reported that Google researchers used SUMO software to build a basic model of Seattle's T-Mobile Park and Lumen Field areas, and used information such as congestion volume, traffic light locations, and average road speeds provided by Google Maps to draw Detailed heat map

▲ Picture source Google official press release (the same below)

The research team then divided the heat map into different areas and introduced a "user behavior model" and route suggestions provided by the Seattle Police Department, thereby establishing a "traffic diversion" that can assign the best route to car owners. "Model

According to a press release from IT House, Google researchers collaborated with the Seattle Department of Transportation in the United States to actually apply traffic diversion artificial intelligence models in multiple large-scale events in August and November 2023, and used "dynamic Guidance Display (Dynamic Message Signs)", the results showed that the congestion time was shortened by 7 minutes on average, and the traffic efficiency was successfully improved by 30%

According to Google, the research shows the potential of "simulation technology" in traffic planning to improve the efficiency of traffic at large event venues and improve the overall traffic environment by allowing road planners to understand underutilized road sections.

The above is the detailed content of Google researches new AI model that can improve traffic efficiency by 30%. For more information, please follow other related articles on the PHP Chinese website!