Multi sensor underwater pipeline tracking with AUVs

The exploitation of deep-water gas and oilfields is increasing. Pipelines transport the produced oil and gas onshore or transfer these between countries through the water. Underwater communication cables on the seafloor are the backbone of our intercontinental communication infrastructure. Sea cables transfer electrical energy from offshore production facilities or between countries and continents. This infrastructure of our economics and society can corrode or can get damaged. The status of these offshore pipelines and cables needs to be monitored regularly and damages need to be found and repaired quickly. This paper presents a method to guide an autonomous underwater vehicle with the information from different sensors including cameras, multibeam echo sounder (MBES), sub bottom profilers and magnetic sensors. Our online sensor fusion and tracking algorithms for underwater pipeline and cable inspection are based on probability maps which contain the location and the rating of each pipeline detection from every sensor. With these algorithms, we want to improve the regular pipeline inspection. The proposed system will be tested for its performance in simulation with different pipeline layout scenarios. Other applications for this map based fusion algorithm (beside pipeline and cable inspection) can be the mapping of toxic substances in the water column and the search of their sources.