Rapid Incident Detection as Easy as 1, 2, 3
Using advanced detection technology for rapid response as well as traffic data collection is now possible. Frances Cook reports.
Iteris is a market leader in the traffic management market and develops technologies that detect traffic incidents to reduce congestion and improve safety, as well as lowering costs incurred by government agencies, commuters and taxpayers involved as a result of traffic-related incidents.
The company has just launched Abacus 2.0, an automated incident detection (AID) software system that is an enhanced version of the company's original Abacus 1 product, which is described by the company as a 'force multiplier'.
It uses existing camera systems to collect real-time traffic data, rapidly identify incidents and gain enhanced traffic flow information in a ready-to-use graphical user interface format.
Using existing cameras for surveillance
The product uses existing surveillance cameras in tunnels, highways and arterial roads to collect traffic data.
It can support up to 12 video feeds, has a web-based graphical user interface for central control and allows for data and video sharing through any web browser.
The system also ensures messages can be delivered immediately to operators as its automated incident detection feature sends incident alerts by email and text message or directly to the traffic management centre's software.
As congestion problems, traffic incident-related accidents and infrastructure costs show no signs of relenting, the need for an effective interoperable incident detection and traffic flow monitoring system is greater than ever.
"The enhancements made to Abacus 2.0 are a logical extension of a product that effectively enables domestic and international agencies to leverage their current surveillance cameras and turn them into real-time data collection and incident detection systems," said Abbas Mohaddes, president and CEO of Iteris.
"We believe this leverage can translate into cost savings for these agencies, which we expect will be a key driver of the product's demand."
According to the Texas Transportation Institute in 2000, the largest 75 metropolitan areas in the US experienced 3.6 billion vehicle-hours of delay resulting in $67.5 billion in lost productivity.
The UK's Department of Transport sees congestion as one of the most serious transport issues, and Rod Eddington's report back in 2006 estimated congestion cost England £22 billion each year.
The most likely places for congestion and accidents to occur are bridges, toll roads and tunnels. "Incidents can include debris, stopped vehicles, wrong-way drivers, pedestrians and bicycles. Debris on the road with vehicles can be a deadly combination with any type of traffic," says Rich Hall, software sales and marketing manager for Iteris in his July 2011 white paper.
On August 2 2011 a 25-ton chunk of concrete suddenly fell from the Montreal's Ville-Marie expressway tunnel onto the roadway.
Fortunately there were not any vehicles using the tunnel at the time, but this shows how fast-working incident detection needs to be when tragedies can happen in a split second, such as the famous Mont Blanc tunnel fire in France on March 24, 1999.
After a Belgian transport truck caught fire in the tunnel, 39 people died and $250 million in damages resulted. Although the cause was disputed, 16 people and companies were tried for manslaughter and the tunnel's head of security was given a six-month jail term. The French company operating the tunnel was given a two-year suspended jail term and a fine of approximately $18,000.
"It is easy to see the need to prevent and respond to tunnel incidents quickly," said Hall. "The evidence is there, that when accidents happen in tunnels, they are serious. AID can help alert officials, allow an operator to visually diagnose the situation, minimise congestion, reduce greenhouse gasses, detour traffic and prevent additional catastrophes resulting from inattention to the initial incident."
Traditionally electric loop detectors have been used to detect vehicles and sense incidents automatically.
According to the US Department of Transportation in 2004 around 20% of all freeway miles in around 48 US metropolitan areas were equipped with electronic loop detectors where the loop information (speed and loop occupancy) was detected by algorithms in software to pick up patterns that demonstrate there is an incident.
However, they do not provide a means of verification because the operator cannot actually see the incident.
"Inductive loops have been used in tunnels to detect the presence and travel of vehicles, but they have become an antiquated technology in recent years and are being increasingly replaced by video detectors.
Video detection sensors have been used for more than 15 years to detect vehicles stopped at a traffic signal, and have become the lower life cycle cost alternative to inductive loops," said Hall.
"Inductive loops require more frequent repair because, cars and heavy trucks travelling over them constantly pound the newly formed edges underneath the sealant created by cutting a channel in the road surface."
Combined incident detection and data collection
The Abacus 2.0 system collects traffic data by using existing video cameras, infrastructure that is already in place, and will also detect incidents, which allows operators and agencies to view the potential danger zone in real time. It offers tracking-based detection algorithms that turn CCTV cameras into a rich source of traffic flow and incident data.
"Only a visual view of an incident can help a traffic management operator determine whether he or she should turn on the fire suppression system, or send firefighters," said Hall.
The system's web-based interface means agencies can view its output from a traffic management centre as well as remote or emergency locations and system automatically monitors a wide range of incidents, including: slow vehicles, stopped vehicles, roadway debris, wrong way drivers and pedestrians.
In terms of its data collection capabilities, Abacus 2.0 effectively collects volume, speed, occupancy, lane-by-lane and vehicle classification data. Costs and time can further be reduced through operators not needing to leave the traffic management centre very often and due to the fact that data is immediately transformed into a configurable GUI reporting interface, reducing analysis time.
"It's easy to see why video sensors and automated incident detection are emerging as the number one choice of traffic engineers for knowing what the true conditions are on a roadway and having the information to act on it. Loops and radar, while they have their place in the ITS industry, cannot tell the operator if the vehicle is engulfed in flames in the middle of a tunnel or if it has a flat tyre, and that kind of information can make a critical difference," said Hall.
"With the ability to increase public safety and reduce the costs of incidents and their resulting clean-up, AID is a win-win for government agencies, commuters and taxpayers."