Public Funding for Cowi UK Limited

The purpose of this project is to develop techniques to reduce the amount of time personnel need to spend on the ground examining tunnel structures and hence reducing exposure to health & safety risks and reducing costs while enhancing asset knowledge of rail staff. This will lead to less disruption to rail passengers by reducing disruptive possessions required for examinations. The project will develop a mobile mapping solution using a combination of technologies. It will build on the road-rail tunnel inspection vehicle developed by Railview as a subcontractor within the IN2TRACK2 project of SHIFT2RAIL and co-financed by Network Rail. This test vehicle builds on the previous Innovate UK INFRAMONIT project. The INFRAMONIT-TUNNEL test vehicle developed incorporates the next generation of Infrastructure Inspection Radar that will scan the tunnel lining and invert to build a 3D visualisation of the asset condition and this will be dedicated to this new project. This technology will be combined with mobile mapping laser scanners and associated positioning equipment to acquire accurate and precise geo-located spatial data as a point cloud representing the surface of the tunnel. The mobile mapping sensor platform will not only capture point cloud data but also 3600 imaging data. This multi-platform approach builds in redundancy in the data and improves survey reliability. The data collected will be processed to build an accurate three-dimensional model which can be viewed and manipulated by engineers. This is a development of current state of the art tools which have been used by COWI for virtual inspections of bridges. A key part of this proposal is to develop the use of machine learning algorithms to process the survey data to automatically detect anomalies and then further, to categories the defects in accordance with the Tunnel Condition Monitoring Index (TCMI). By adding intelligence to defects and using state of the art survey equipment, data from future surveys can be compared with the baseline, quickly identifying changes since the previous survey. A combination of semi-automated defect detection and manual virtual inspections will be used to allocate the TCMI codes to defects and generate a report in the required format. The scanning and modelling process will result in much richer background data than that contained in the standard inspection report, which can then be used as part of the ongoing asset management process, including specification of remedial measures. The final stage of the project will be to undertake an Operational Environment Demonstration and evaluate the results in conjunction with Network Rail's asset managers. The key innovations of this proposal are to develop a new work flow for tunnel examination including: application of subsurface inspection radar technologies to the tunnel environment; improvements in positional accuracy; machine learning to automate the identification of certain defects; and production of a data-rich three-dimensional model of the tunnel to facilitate ongoing asset management and maintenance activities.

To develop a novel low-cost and low carbon technique for tunnel face stabilisation