Positioning accuracy of an automatic scanning system for GPR measurements on concrete structures

Most GPR measurements use hand-held survey wheels for positioning, something not feasible in some scenarios, such as measuring a tall retaining wall, tunnels or if tedious and repetitive measurements are required. For situations like these an automated positioning system is required and was therefore developed at BAM. Such a system was tested to calibrate the positioning of a GPR grid measurement which was performed in regular one-direction traverse mode and in a meandric mode. The preliminary results show a systematic error in the positioning, which depends on the traveling speed of the scanning system, because there is an electronic delay of the trigger signal. These errors are duplicated in meandric mode compared to the errors occurring in one-direction traverse mode. This may be insignificant when larger objects are investigated, but can be important for the identification and location of small objects which show weaker reflections, like e.g. compacting faults with a size of a honeycomb.