Some Issues on the Design of a Low-Cost Autonomous Underwater Vehicle with an Intelligent Dynamic Mission Planner for Pipeline and Cable Tracking

G. G. ACOSTA; H. J. CURTI; O. A. CALVO; S. R. ROSSI

Underwater Vehicles

In-Tech

Lugar: Viena; Año: 2009; p. 1 - 18

The main goal of this chapter is to introduce to the design and working hypotheses for the construction of a low-cost Autonomous Underwater Vehicle (AUV) devoted to pipeline and cable inspections. Underwater inspection is mainly necessary on account of the periodic surveys for the preventive maintenance of submarine infrastructure in the off-shore industries. The advantages of doing them with AUVs instead of Remote Operated Vehicles (ROV) or Towed Unmanned Devices (TUD) are low costs and better data quality in the inspection missions. The vehicle presented in this chapter was thought as an experimental platform to test multiple algorithms and to develop new technology, mainly using artificial intelligence techniques. In this sense, its modular architecture tackles the four main needs that should be solved in an autonomous vehicle to perform useful tasks in an unknown and strongly disturbing environment like the underwater world. They are: a) a robust control system to manage nonlinearities and disturbances, b) a precise guidance system to avoid unnecessary time and thus energy consumption, c) an accurate navigation system to determine self and target’s positions, and d) an intelligent dynamic planner proposing the best possible trajectories and actions to successfully reach the mission objectives, based on decisions taken without human intervention. In a comparison with a biological being, the autonomous robot also need some kind of controlled muscles and forces to move, self perception and notion of the surroundings, and a brain to plan actions and movements. In this work, different approaches for all of the aforementioned systems will be presented and thoroughly analyzed at the light of experimental evidence and author’s experience in mobile robots (Fernández León et al., 2008). These experiments comprise computer simulations, hardware in the loop simulation as well as sea trials with the low-cost prototype described in the sequel, which is expected to navigate in the sea up to 100m of depth.