INVESTIGADORES
ROZENFELD Alejandro Fabian
congresos y reuniones científicas
Título:
SMOOTH PATH PLANNING FOR AUTONOMOUS PIPELINE INSPECTIONS
Autor/es:
CALVO IBÁÑEZ, O.A.; SOUSA, A.L.; ROZENFELD, A.F.; ACOSTA, G.G.
Lugar:
Bremen
Reunión:
Conferencia; 6th International Multi-Conference on Systems, Signals and Devices; 2009
Resumen:
This paper details a control method, used to guide an autonomousunderwater vehicle (AUV) in the task of locatingand tracking underwater pipeline and cables. The articledetails the simulation and experimental results of thenavigation when searching for simulated pipes on the seabedin shallow waters.Sonar based pipeline detection and tracking starts bythe acquisition of the sonar image and further realtimeprocessing to distinguish the pipe from the seabed. Dependingof the success of detection an AI based dynamicpath planner must decide the next actions: a search or atrack process. Pipeline searching involves a series of zigzag explorations of the area where the pipe is presumablylocated. Pipe tracking involves keeping the navigation atfixed distance from the pipe. The path following methodproposed in this article is based on a combination of Lyapunovtechnique and PI controller acting over two horizontalthrusters dual torpedo AUV.The article details some of the stages involved in thedetection of the pipe using sonar, the control method usedfor smooth path following, the main aspects of dynamicmission re-planning, and the results, both simulated andexperimental for search process of a simulated pipe in shallowwaters.