University of Minnesota Driven to Discover
U of MNUniversity of Minnesota
Center for Transportation Studies

Programs & Labs

ITS Sensor Header

Winter 2009/2010

A portable weigh station for rural roads

Taek Mu Kwon

Modern engineering research has quantified the link between axle weight and road wear: a linear increase in loading produces a fourth-power exponential acceleration in wear, so doubling axle weight is estimated to accelerate road wear by roughly 16 times the original rate. This calculation is a key component of pavement design and maintenance planning.

Today, however, the large number of commercial vehicles moving at speeds no Roman oxcart driver could have imagined has made the problem of regulating vehicle weights extremely challenging. Transportation agencies around the world are turning to new technologies to help monitor the impact of heavy vehicles on valuable pavements.

Professor Taek Mu Kwon of the University of Minnesota–Duluth electrical and computer engineering department is developing a portable weigh-in-motion (WIM) system that will enable highway engineers to measure truck weights in areas far from highway weigh stations.

Kwon is also the director of the Transportation Data Research Laboratory. His past research has included developing novel portable, wireless, self-powered vehicle detector systems as well as managing the archiving and distribution of traffic data from the Twin Cities freeway system.

Highway engineers charged with maintaining rural roads have noted more heavy vehicles in recent years. The increase, according to researchers, is being driven by the rapid development of the biofuels industry and its need for large quantities of corn and soybeans.

The potential for increased heavy truck traffic to damage local roads concerns highway engineers in rural areas. Estimating road wear due to heavy trucks is difficult in rural areas, however, because constructing traditional weigh stations in these areas is unfeasible. WIM technology has the potential to monitor truck loads more economically. Kwon is designing a system that is portable and can be deployed where it is needed to monitor vehicle loads.

Traditionally, weigh stations staffed by trained personnel have been the primary tool for determining compliance with weight limits. After a vehicle has been stopped and weighed, weigh station staff determine whether its weight is within the limits prescribed for its classification and any applicable permits.

Constructing and maintaining weigh stations on low-volume rural roads, however, is not economically feasible. A portable WIM system gives engineers the flexibility to deploy monitoring equipment quickly and change locations as necessary, which makes monitoring vehicle weights in multiple locations cost-effective.

Weigh stations have several other disadvantages. Requiring commercial vehicles to stop for inspection can cause significant delays, especially on heavily traveled routes where many vehicles may be forced to wait their turn on the scales. Weigh-in-motion systems avoid these drawbacks by replacing staffed weigh stations with automated equipment. Instead of pulling off the highway to be inspected, vehicles simply drive over WIM scales at a constant speed. Weight sensors can be combined with automated vehicle classification systems and identification via license plate recognition or other means, enabling direct communication of violations to enforcement personnel.

One of the known problems of permanent WIM systems is that calibration is laborious and difficult. A WIM system calibration method recommended by the FHWA requires trucks of known weights to make multiple passes over the sensors so that the statistical average of recorded weights can be calibrated against the known weights. Unfortunately, weight measurements by permanent WIM stations can be affected not only by the truck’s actual weight but also by the pavement surface conditions and winds. In addition, the calibration truck must run along with regular traffic, so engineers must manually insert markers into the data to designate passes by the calibration truck, which is cumbersome.

Kwon’s solution is to place the portable WIM side-by-side with a permanent WIM station and collect data over a longer period under the same environment, instead of over only a few runs. Since a portable WIM station can be calibrated in a controllable environment such as a parking lot using a vehicle with a known weight, it is much easier to calibrate the portable system.

To develop a portable WIM system, several technical obstacles had to be overcome. First, the sensor pads needed to be small and light enough to be moved to remote locations while remaining strong enough to withstand being driven over by heavy trucks.

The heart of Kwon’s WIM design is a pressure sensor consisting of a thin strip of piezoelectric material, which converts mechanical pressure into a measurable electrical signal. 

To protect the piezoelectric beam, Kwon examined a variety of flexible materials for the weigh pad. Cementing together layers of neoprene fabric and ballistic nylon (a fabric used in bulletproof vests) to create a flexible pad proved problematic due to the difficulty of bonding the layers together without causing the pad to warp. He then turned to industrial reinforced-rubber conveyor-belt material. A groove is carved into one layer of belt material to hold the piezoelectric beam, and a second layer is cemented over the top to fully enclose and protect the sensor.

A second obstacle was maintaining accuracy under tough real-world conditions. The signal processing system must be capable of measuring axle weights while isolating peripheral forces from the main load force. The impact of heavy wheels can cause anchored sensor pads to shift slightly, resulting in erroneous measurements. Temperature variations can also interfere with measurement.

Finally, to be deployable in remote areas that lack electrical service, the system must run entirely on battery power and be able to operate continuously for long periods. Because calculating vehicle weights from raw sensor data is computationally intensive, the data processor must be powerful but draw little electricity.

The system design includes a signal conditioning circuit to prepare raw analog signals from the sensor pad for digital processing. This critical circuit minimizes electrical impedance mismatch, removes noise in the signal caused by thermal fluctuations, and maps the raw charge signals onto linearly proportional voltage signals for digitization. The conditioning circuit will incorporate specially designed amplifiers tuned to the characteristics of the input signal, currently under development in Kwon’s laboratory.

Following the conditioning circuit, final processing of signals from the sensor pad will be accomplished by software that includes modules for real-time plotting, digital data conversion, vehicle segmentation, axle spacing and weight computation, and vehicle classification. The primary focus of software development will be on a vehicle segmentation algorithm based on axle spacing and weight.

Development of the portable WIM system has been made easier by a hardware-in-loop (HIL) testing system developed by Kwon in a previous WIM research project. The HIL system allows developers to run an unlimited number of signal tests in the laboratory using analog signal inputs controlled by software. The HIL system can also produce erroneous signals to test the WIM device’s handling of errors.

Field testing of the prototype WIM system is scheduled to take place during 2011 at the Minnesota Department of Transportation’s MnROAD pavement testing facility, where vehicles of known weights can be operated on a variety of pavement types to conduct controlled verification of the system.