Characterization and mitigation of temperature sensitivity within mobile retroreflectometer unit

Mobile retroreflectometers have demonstrated great potential as a tool to assess the retroreflectivity of pavement markings, on both project and network levels. Such devices can collect and process much more data than their handheld predecessors can. However, studies have shown concerns with repeatability and reproducibility. Successful integration of pavement marking retroreflectivity readings into a pavement management system hinges on the accuracy of the collected data. Previous research has centered on data collection techniques to minimize error (collection frequency, averaging), but little research has focused on the technology and the components making measurement possible. Results were summarized for the characterization and the mitigation of a temperature sensitivity uncovered during the Florida Department of Transportation's evaluation of a mobile retroreflectometer unit. During routine data collection, offsets in the measured retroreflectivity were observed between results from repeat runs on the same section. A strong correlation was observed between the temperature of the device and measured retroreflectivity. More specifically, as the temperature within the device increased from 20 degrees C to 40 degrees C, a 30% increase in measured retroreffectivity resulted. Testing revealed that the optical filters operating outside their manufacturer-specified temperature range were primarily responsible for the temperature sensitivity. The response of the filter set over a broad range of temperatures was characterized, and a compensation algorithm was created to reduce system error. A thermoelectric cooler was incorporated to regulate the internal temperature of the retroreflectometer. With these modifications, a temperature sensitivity of +/-3% resulted over a temperature range of 20 degrees C.