AeroInspekt

Automated Measurement of Rail Infrastructure in the Hamburg Harbour

In the CTA terminal run by the Hamburger Hafen und Logistik AG (HHLA), containers are transshipped mostly automated. The rails used by the automatic gantry cranes must be checked regularly for deformations and measured with millimeter precision. The current methods to do this are cost intensive as the cranes have to be stopped for several hours.

AeroInspekt aims to measure the rails using a multicopter without interfering the operation of the terminal. The drone is taking highly overlapping aerial images of the rail areas and calculate a three-dimensional model out of them by photogrammetry.

Automated reactive flight planning

To measure the rails with millimeter precision the flight system takes aerial images with a ground resolution of under 1 mm. To map the whole measurement area, unobstructed images of the rails are needed. Significant parts of the rails will normally be covered by the operating cranes, leading to a loss of relevant information in the mapping result.

The ground control app Inspekt GS developped in AeroInspekt solves this problem with the implementation of an automated reactive flight planning function. It recognizes relevant obstrucctions with a machine learning approach while still in flight. The accordingly marked sections are covered again until sufficient data is gathered.

Automated Analysis

To connect the three-dimensional data to the harbour reference system with millimeter precision, specifically designed hybrid markers are installed in the track bed. With a combination of prisms and machine-processable optical markers these can be both measured from the ground with a total station and recognized automatically in the aerial images. After the flight, the raw data is processed with a photogrammetry workflow to create a three-dimensional point cloud.

To extract the rails position from the point cloud, we created the program Inspekt XT. It detects the rails profile in periodically generated two-dimensional slices and calculateds the rails position. The issued measurement report can by used by the responsible authorities to recognize a breach of thresholds that might disturb the operation and take appropriate actions.

Publications

• Ghassoun, Y.; Gerke, M.; Khedar, Y.; Backhaus, J.; Bobbe, M.; Meissner, H.; Tiwary, P. K.; Heyen, R. Implementation and Validation of a High Accuracy UAV-Photogrammetry Based Rail Track Inspection System. Remote Sens. 2021, 13 (3), 384. https://doi.org/10.3390/rs13030384.