SmartPark Truck Parking Availability System: Magnetometer Technology Field Operational Test Results

The purpose of this project was to conduct field operations tests to determine the feasibility of using magnetometry to determine truck parking availability. The field operations tests were conducted at two locations: a private truck stop, Interstate Travel Plaza, on U.S. 1 in Wrentham, MA and at a public rest stop, Mile Marker 9 on I-95 northbound in Mansfield, MA. The magnetometry technology detects the presence of a truck by a disturbance in the earth’s magnetic field. The technology deploys an inexpensive wireless, battery-powered vehicle sensor unit (size of a soda can) embedded in the pavement to measure the disturbance and trigger the event. A base station unit wirelessly collects the event data from a group of four to six vehicle sensor units. A parking area relay wirelessly collects the event data from all the base station units and transmits the aggregated data to a centralized database server where a detection/classification processor uses rules-based algorithms to determine truck parking availability for an entire truck parking area. A ground truth camera was installed to allow for manual verification of (1) the detection of a vehicle (presence or absence) and (2) the classification of a vehicle (car, truck, or other). The magnetometry performed vehicle detection at an accuracy rate of 96.2% at Mile Marker 9 (n=2056 vehicles) and 96.5% at Interstate Travel Plaza (n=1241 vehicles). There were no overcounts, but there were undercounts (false negative detections). Furthermore, the magnetometry achieved only a vehicle classification accuracy of 92.2% at Mile Marker 9 (same sample size as given above) and 78.6% at Interstate Travel Plaza (same sample size as given above). Because the magnetometry technology did not meet performance requirements for vehicle classification accuracy, FMCSA will be repeating the field operations test with another technology to be announced.