IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Las Vegas Analyzes Traffic Near-Misses with Edge Computing

Las Vegas plans to expand a traffic analysis project across downtown to gather data related to drivers heading the wrong direction on one-way streets, among other findings.

In a test to learn just how many drivers would head the wrong way on a one-way street in downtown Las Vegas tech officials installed a sensing system which would monitor and analyze the street’s traffic 24 hours a day. They expected to learn maybe one or two cars a week would drive the wrong direction.
 
“It was 30 to 40 cars per week, going the wrong way,” said Michael Sherwood, director of Information Technology for Las Vegas.
 
The city then installed additional signage to see if this might correct the problem. The number of wrong-way drivers decreased, but not as much as “we would have liked to have seen,” said Sherwood. “But it was a reduction.”
 
The small, one-way-street test site was part of a pilot project to explore how collecting and analyzing traffic data can inform decisions around how to best respond to safety issues, as well as learn more about traffic patterns. With technology from NTT, a Japanese technology firm, and Dell, the city has grown the pilot to 16 intersections and plans to roll out the system across the core of downtown Las Vegas. Some of the most important data transportation officials want more insight into is the number of “near misses,” where autos nearly encounter other cars or even pedestrians.
 
“When you have all these one-way streets, and there’s an accident, generally, there’s a knee-jerk reaction: ‘we have to do something on that street,’” said Sherwood. “That might have been the street that had the least number of wrong-way drivers the whole year.”
 
While other streets maybe have not yet had a serious accident, but yet saw a number of near-misses, said Sherwood.  
 
“So we’re using edge analytics now to monitor one-way traffic,” said Sherwood.
 
The technology uses infrared cameras and LiDAR, a radar system often included in autonomous vehicle technologies, to distinguish cars from pedestrians, as well as take in other measurements around speed and direction. Out of concerns for privacy, the technology does not read identifying features like faces or license plates, said Sherwood.
 
“For us, it’s more about the data… We’re not interested in the individual. We’re interested in the outcome,” he added. “We’re interested in, did someone go the wrong way on a street? And how many times is it happening?” (The pilot did use camera technology to allow the human analysts to verify the accuracy of the overall system.)
 
This detailed level of street analytics can be put to use not only by public works teams exploring where to best install one-way signage, but also maintenance crews needing to know when’s the best time schedule road-closures, or economic development analysts wanting detailed data related to car, bike or pedestrian traffic.
 
“If we’re able to reduce the number of wrong-way drivers, now we’ve reduced the number of paramedics and police-calls,” said Sherwood, pointing out one of the end results that can come from analytics. “Because if we reduce or eliminate accidents, now we have more public safety services available to handle other emergencies, in a more timely fashion.”
 
But its not just the collection and analysis of large amounts of traffic data the technology is enabling. The edge-computing component means that data can be analyzed and acted upon in real-time.
 
“So now you’re able with edge analytics to not just get data and look at a dashboard and compute… But is it possible to use advanced computing technology to affect outcomes, and prevent accidents, even when someone might make a wrong decision,” said Sherwood.
 
Edge computing and analytics also paves the way for supporting autonomous vehicles, technology which Las Vegas has been a leader in with projects like the low-speed autonomous shuttle route in downtown.
 
“By having all the data at the edge, now we’re able to start some ground-breaking work, we believe,” Sherwood added. “And so how can we now transmit that to an autonomous vehicle. So that now the vehicle has the information before it even gets to the intersection.”
 
By using edge computing, the city is able to make decisions faster and alter the environment in real time. For example, a person detected in a crosswalk to trigger a longer signal light sequence.
AVs could make decisions related to speed, stopping and other driving metrics based on roadway data coming into the system from edge computing analytics.
 
“At the end of the day, for us, it’s no longer about being called a ‘smart city,’” Sherwood remarked. “It’s about being a ‘connected community.’”
 
Skip Descant writes about smart cities, the Internet of Things, transportation and other areas. He spent more than 12 years reporting for daily newspapers in Mississippi, Arkansas, Louisiana and California. He lives in downtown Yreka, Calif.