Engineers at Rutgers University in New Jersey have combined ultrasonic sensors and GPS receivers to create a low-cost, efficient way to find the nearest available parking space.
"The goal here is to come up with a way to figure out where parking is in a timely manner and also to get statistics about which areas are most crowded so that the pricing of the meters can be done right," says Suhas Mathur, lead student on the project.
After doing extensive research on traffic congestion in downtown areas, the team from Wireless Information Network Laboratory (WINLAB) decided to look specifically at the issue of parking. Besides reducing traffic jams, better access to parking can reduce wasted hours and excess CO2 emissions that result from motorists roaming streets looking for parking spots.
The WINLAB team collected data from ultrasonic sensors mounted on passenger-side doors of three vehicles that travelled the streets of New Jersey for a two-month period. The low-cost ultrasonic sensor range finder combined with a GPS system allowed the team to detect obstacles or parked cars in roadside parking spots, as well as free spaces.
The sensor "measures the distance to the next obstacle at the side of car. And if a car is driving in the right-most lane, then if cars are parked at the roadside, the sensor will register a small distance and if there is no car parked, it will register a larger distance," explains Marco Gruteser, assistant professor at Rutgers University.
During the two-month test period, the scanning vehicles were used to analyze two different types of parking in downtown areas.
Mathur says they looked first at streets with meters or slotted parking. By counting the number of spots occupied in relation to the total number of available spots, the team's test vehicle was able to locate available parking and give the researchers an idea of parking patterns for different areas of the city.
The team also looked at areas where people were allowed to park along the sides of streets, but where there were no meters or marked roadside slots. They studied the number of vehicles that could fit on a given street, based on space between two vehicles, when there were no designated slots.
Gruteser hopes to employ this system in a larger or real-world setting by mounting sensors on municipal vehicles and taxicabs.
He says he hopes the findings could be shared with maps like those offered by Google to give drivers information about specific streets where parking is easier to find. A network of sensor-equipped vehicles could provide precise real-time information on how many spots are available in a given area.
For parking updates every 10 to 20 minutes in a city like San Francisco, for example, the team estimates it would need about 300 equipped taxicabs and a $200,000 US budget per year for initial installation and operation costs.
This investment could be offset by the economic benefits of streamlining traffic flow, especially relevant for big cities like Toronto.
A report from October 2009 by the Organization for Economic Co-operation and Development says Toronto's traffic problems cost an estimated $3.3 billion in lost productivity annually because of traffic congestion on streets and highways, coupled with the growth problems associated with Toronto's public transit system.
According to Transport Canada, Toronto has the highest annual total cost of congestion of any Canadian city.