Sparse representation for artefact/defect localization with an acoustic array on a mobile pipe inspection robot

Autonomous robots equipped with long-range acoustic sensing are revolutionizing the inspection of extensive sewer networks by accurately detecting and mapping artefacts and defects that could cause a pipe operational failure. This study introduces a new frequency domain Sparse Representation (SR) method for identifying artefacts and defects using a circular microphone array mounted on a continuously moving inspection robot. The circular array enhances the precision of localization by extracting plane wave modes at frequencies above the pipe's first cut-off frequency. The proposed mobile compensation algorithm can separate overlapping echoes with less than 0.5% error of the sensing distance, given sufficient overlap in acoustic wavelengths. It also compensates for phase shifts and reduces noise from a moving robot, enabling accurate estimation of defect direction and distance amid overlapping acoustic echoes. The novelty of this method is in the ability to work on a continuously moving robot to achieve a significant enhancement in terms of operational speed, direction estimation and defect resolution quality.