Printer-friendly version

Traffic Surveillance > Road-based Technologies > Roadside >

Other Roadside Detectors

• Introduction
  • Active/Passive Infrared Detectors
  • Microwave
  • Radar
  • Ultrasonic
  • Passive Acoustic
• Case Study
• References

INTRODUCTION

Roadside (or non-intrusive) detector systems are increasing in prominence due to today’s congested freeways and signalized intersections. These technologies cause minimal disruption to normal traffic operations and do not need to be installed in or on the pavement. They are generally mounted overhead or to the side of the pavement, often on pre-existing structures. Roadside detection technologies are usually used to count traffic volume and classify vehicles. Typically, there are 2 types of non-intrusive sensors: passive and active. Passive sensors do not transmit energy, but rather detect energy that is emitted by the vehicles and roadway or is reflected from them. Active sensors transmit energy, a portion of which is reflected or scattered from the vehicle and roadway back toward the sensor. Most of these sensors are cost competitive with inductive loop detectors, if not cheaper. See our Telecommunications Diagram on Roadside Sensors for more information.

Active/Passive Infrared Detectors

Active Laser infrared Detectors
Active infrared (or laser range finder technology) is used in a wide range of applications, including measuring vehicle emissions, military targeting, aircraft obstacle avoidance, and spacecraft docking. Laser detectors operate on the same principles as microwave radar but transmit energy at higher frequencies (shorter wavelengths). The detector senses a portion of the reflected energy in its field of view. The distance of an object from the detector is found by measuring the two-way travel time of the infrared pulse, from the detector to the target and back. Lasers provide presence, speed, volume, occupancy, and classification information in day and night conditions. IR detectors are vulnerable to weather conditions such as fog, clouds, shadows, mist, rain, and snow which scatter and attenuate wave energy. Problems also may arise when radar locks onto and measures the speed of the strongest backscattered signal, excluding smaller vehicles in the same area. Another downside to infrared detectors is their high cost, particularly at intersections. The active detectors are more expensive than the passive.

Cost per Lane: $1,293 (1999)

Cost per Lane: $443 (1999)

Microwave

Microwave detectors fall in two categories: Doppler or radar devices. Radar devices, also known as pulse microwave, measure the time it takes for a portion of the microwave radiation to be reflected from the target area to a receiver. Microwave radar vehicle detectors transmit electromagnetic energy at the speed of light in frequency bands between 2.5 to 24.0 GHz. They are able to count vehicles, measure speed and detect vehicle presence. Experimental models have been used for vehicle classification by measuring the vertical profile of a vehicle. They are generally insensitive to weather, provide day and night operation, and perform best on fairly open roads where long-range capabilities can be taken advantage of. Advanced units can measure target distance, thus one unit can have multiple detection zones.

Doppler devices, also known as continuous microwave devices, output a continuous signal to the detection zone and use the Doppler principle to analyze the change in frequency of the reflected signal to calculate the speed of the vehicle. Stationary vehicles can not be detected with this wave form. The frequency-modulated, continuous wave signal, or radar devices, allows the measurement of both speed and presence, as well as the detection of stationary vehicles. Pulse waveforms are used with ultrasonic and laser radar sensor technologies. These technologies measure distances to the road and vehicles, providing vehicle count, presence, and occupancy information. By transmitting pulse energy at two calibrated, closely spaced incident angles, vehicular speed may be measured by recording the time at which the vehicle crosses each beam.

Microwave systems provide a cost-effective alternative to loops for vehicle presence detection. They are small, lightweight and easier to install than loops, and they can operate over a long range. Their size, low cost and low power consumption makes them well suited for surveillance both at intersections and on freeways. One concern when installing microwave systems is the potential for interference since there are many applications for microwave technology in modern society.

Cost per Lane: $659 (1999)

Radar

Radar detectors are in limited use in incident detection and freeway management projects. The most common type of radar directly measures vehicle speed using the Doppler effect (measuring frequency shifts between the transmitted and received beam caused by the vehicle motion). Vehicle counts can be determined by accumulating each vehicle detected, but this approach cannot readily obtain lane occupancy and vehicle lengths. Similarly, detection of stopped vehicles or very slowly moving vehicles is difficult. A second type of radar detector transmits a signal that is swept over a range of frequencies. This technique measures the range to the vehicle and thus can detect presence.

Radar detectors of the Doppler and swept frequency variety require one antenna per lane, mounted on a structure or a sign bridge over the lane. If such a mounting arrangement is not available, the installation cost is significant when the sign bridge is included. The IVHS Detection Technology project early results show that these radar detectors have an accuracy within the range from +0.5% to +6%.

Another type of radar detector can be mounted at the side of the road and scan up to twelve lanes from one location. Since light poles and utility poles are often available, or can be readily installed, the device is more cost effective. The device can also detect vehicle presence, and is thus able to determine occupancy and existence of stopped vehicles. However, it does not measure speed directly, relying upon the standard "single loop" speed estimation equation based upon average vehicle length. The accuracy of this device is in the +5% range.

The advantages of radar devices are that it is easy to use and requires no cutting of pavement or disruption of traffic flow for installation or maintenance (if mounted on a structure or sign bridge where overhead access is possible). For the Doppler units, direct speed measurement is a significant benefit. Similarly, when traffic lanes are relocated, radar antennas can be easily re-aimed. The disadvantages of radar are the requirement for a structure or sign bridge for overhead mounting, the limited field operational experience, the small number of vendors in the market, and the difficulties of accurately sensing lane occupancy and slow moving or stopped vehicles with the Doppler units.

Cost per Lane: $314

Ultrasonic

Because ultrasonic waves propagate through the air, they are subject to attenuation and distortion from a number of environmental factors including changes in ambient temperature, air turbulence, and humidity. Nearly all targets reflect ultrasonic sound waves, but textured or porous surfaces produce a weaker echo.

Pulsed Ultrasound
Vehicle profiling can be achieved by installing a pulsed ultrasonic detector above the roadway. The vertically aligned (downward looking) transducer measures wave travel time to the pavement or to the top of passing vehicles. A fast pulse repetition rate (~13 Hz) allows a minimum of 2 to 3 measurements with up to 1" resolution to be made. Excellent classification performance can be achieved for most vehicle types, though the sensor can have difficulty differentiating between cars and vans. Air turbulence and temperature adversely affect performance.

Ultrasound detectors have no moving parts so they tend to be reliable, durable and require little maintenance. They are also small and can be sited permanently or used as a portable unit.

Cost per Lane: $644 (1999)

Passive Acoustic

This detector is usually configured as a vertical dipole array of microphones, which "listen" to noise produced by approaching vehicles. The time delay between the arrival of sound at the upper and lower microphones changes with time as the vehicle emitting the sound passes under it. When the vehicle is distant from the microphones the sound arrives almost instantaneously at both phones. When the vehicle is under the microphones, sound reception at the upper microphone is delayed by the intersensor distance. Vehicles are tracked using cross-correlation between phones, and the best results are achieved when the data is filtered to a bandwidth of 50-2000Hz. In this band frequency content includes both engine and tire noise (though most of the acoustic noise is produced by the tires). Interference between the noise of multiple vehicles is a key limitation to acoustic technology. Performance is also affected by low temperatures, snow and very dense fog which can muffle sound and lead to undercounting.

Cost per Lane: $486 (1999)

CASE STUDY

Minnesota Guidestar Program

System Description:

1. Passive Infrared
2. Active Infrared
3. Passive Magnetic
4. Radar
5. Doppler Microwave
6. Pulse Ultrasonic
7. Passive Acoustic
8. Video

Evaluation of each technology's data collection capabilities covered both the quality and types of data collected. Emphasis was placed on urban traffic conditions, such as heavy congestion, and locations that typify temporary counting locations, such as 48-hour or peak hour counts. The evaluation also focused on the ease of system setup and use, general system reliability, and system flexibility. The performance of the technologies was evaluated under the extreme weather conditions found in Minnesota over the year. The technologies were evaluated at both freeway and intersection locations to provide a variety of traffic conditions.

Data were collected in both 24-hour and multiple week continuous test periods. When evaluating the various technologies it is important to examine the specific test conditions because the test periods included multiple mounting locations and test conditions over the year.

Conclusions:

• Most of the devices tested in this project are well-suited for temporary counting situations. Ease of installation and flexibility in mounting locations and power supplies are important elements in selecting a device to install quickly and move from location to location.
• The devices that use Doppler microwave, active infrared, and passive infrared technologies have a simple "point-and-shoot" type of setup.
• Passive magnetic, radar, passive acoustic and pulse ultrasonic devices require some type of adjustment once the device is mounted. In most cases this adjustment must be performed over a serial communication line.
• Video devices require extensive calibration over serial communication lines and are not well-suited for temporary counting.
• Extensive installation work is required for video and passive magnetic devices, making them less suitable for temporary data collection.
• From an overhead mounting location at the freeway test site, the video and passive acoustic devices have been found to count to between 4 and 10 percent of baseline volume data.
• Pulse ultrasonic, Doppler microwave, radar, passive magnetic, passive infrared, and active infrared have been found to count within 3 percent of baseline volume data.
• The count results are more varied at the intersection test site. The pulse ultrasonic, passive acoustic, and video devices were generally within 10 percent of baseline volume data while one of the passive infrared device was within 5 percent.
• Speed data were collected from active infrared, passive magnetic, radar, Doppler microwave, passive acoustic and video devices at the freeway test site. In general, all of the devices were within 8 percent of the baseline data. Radar, Doppler microwave, and video were the most accurate technologies at measuring vehicle speeds.
• Video and radar devices have the advantage of multiple-lane detection from a single unit. Video has the additional advantage of providing a view of the traffic operations at the test site.
• Weather variables were found to have minimal direct affect on device performance, but snow on the roadway caused some vehicles to track outside of their normal driving patterns, affecting devices with narrow detection zones.
• Lighting conditions were observed to affect some of the video devices, particularly in the transition from day to night.
• Extremely cold weather made access to devices difficult, especially for the magnetic probes installed under the pavement.
• Urban traffic conditions, including heavy congestion, were found to have little affect on the device performance.

There are ongoing developments in non-intrusive vehicle detection technologies. Devices are now available that incorporate multiple technologies within a single device. Developments in other technologies, such as passive millimeter microwave and infrared video, will produce additional entries into the market. At the same time, existing technologies are continually being improved upon.

REFERENCES

Bahler, Stephanie. Kranig, James. Field Test of Non-Intrusive Detection Technologies. Office of Advanced Transportation Systems, Minnesota Department of Transportation, Minneapolis, MN, 1998.

Detection Technology for IVHS: Volume 1, Final Report Addendum. FHWA, 1995. Link to Report

Evaluation of Non-Intrusive Technologies for Traffic Detection. Minnesota Department of Transportation, Office of Traffic Engineering/ITS Section, October 2001. Link to Report

Field Test of Monitoring of Urban Vehicle Operations of Non-Intrusive Technologies. US DOT, FHWA, 1997. Link to Report

Freeway Management Handbook

Middleton, Dan. Jasek, Debbie. Parker, Ricky. Evaluation of Some Existing Technologies for Vehicle Detection. Texas Transportation Institute, Austin, Texas, 1998.

Steinbach, Bertrand. Non-Intrusive Traffic Detectors. ASIM Technologies, Ltd., 1998. Link to Report

Author: Dimitri Loukakos.  Last update: 12/20/01