Automating the evaluation of urban roadside drainage systems using mobile lidar data.

Roadside channel systems are critical for the management of stormwater runoff and the protection of the structural integrity of roads; and thus, require systematic evaluation and maintenance. However, the evaluation of these systems remains ad hoc due to the lack of efficient inspection methods. This paper contributes to filling this gap by providing an automated process for the inspection and evaluation of roadside channel systems using data obtained from mobile lidar (Light Detection and Ranging) scanners. The Cloth Simulation Filtering algorithm was employed to split lidar point clouds into bare earth and object datasets, and then compute six key attributes of roadside channels based on the bare earth dataset. The six attributes are: channel depth, bottom width, side slope, longitudinal slope, and length and density of subsurface pipes and culverts. To test and demonstrate the new automated process, it was applied to six street sections in the City of Houston, Texas. The computed attributes were compared to the design and maintenance guidelines set by the City of Houston and Harris County. The evaluation results indicate that every channel examined in this study has its own condition issues and improvement needs. While no channel section in this study was in full compliance with the guidelines, no channel was utterly incompatible either. The results show that the developed automated process can effectively and efficiently evaluate roadside channels, providing an alternative to conventional manual inspection methods. Future work could use the output of the developed method to assess the risk of localized flooding and inform both the municipal government and property owners about effective mitigation measures. • A method for automating the evaluation of roadside drainage channels in urban settings has been developed and demonstrated. • The automated evaluation method consists of computing six key attributes of roadside channels from mobile lidar raw data. • The potential for evaluating roadside channels at shorter distances to inform targeted improvements is presented. [ABSTRACT FROM AUTHOR]