Knowledge CentreTechnical Resources SearchConference PapersAccuracy Limits of Geometric Analysis Based on Data Collection Vehicles

Accuracy Limits of Geometric Analysis Based on Data Collection Vehicles


Many techniques are currently available for measurement and interpretation of the geometric features of the roadway. Traffic speed, pavement survey vehicles such as the ARAN, are readily equipped with highly accurate location referencing equipment, and have proven to be a time and cost effective means of collecting horizontal and vertical curvature parameters. The accuracy of a survey vehicle’s combined Geographical Positioning System (GPS) and Inertial Navigation System (INS) allow for the very effective measurement of the driven path (reproduction of latitude, longitude, elevation and heading) to be collected at traffic speed, and under a range of satellite interference conditions. The real world accuracy of an ARAN with combined GPS and INS system allows for a trail with a positional accuracy of under 0.5 m to be reported. Curve-fitting algorithms, used to post process collected position data, is highly capable of interpreting the location coordinates into paths of correctly assigned tangent sections and circular curves / parabolas along horizontal and vertical alignments. Using traffic speed, survey vehicles for the collection of horizontal alignment data, factors such as vehicle wander within lanes, lane changes, and lane closures have the potential to affect the integrity of geometric data. The level of error induced by these movements is very dependent on the arc radius and arc length. For shorter arc lengths, the potential curve radius error from vehicle wander (in excess of 5% for small radii curves) becomes more of concern as curve deflection increases. Consequently, and when considering the sensitivity of crash frequency based on the percent error in the prediction of curve radius, just care is recommended when applying design restrictions, such as minimum radius and design speed, for curves of shorter curve arcs. This curve alignment process of combining data collected by traffic speed, survey vehicles with curve fitting post-processors, has been found to be very effective means of measuring a range of geometric policies. When planning roadway geometric surveys it is important to understand the purpose and accuracy level necessary for the different data uses. 

Conference Paper Details

Session title:
Andrew Howard
D.J. Swan
Geometric design