Guidance algorithms for UUVs obstacle avoidance systems
R. Moitie, Nicolas Seube · 2002
To widen the range of application of underwater vehicles both in industrial application and in environmental survey, we need to design powerful robotic systems, exhibiting a maximum degree of autonomy. Ability to construct a reliable representation of the underwater environment, to plan viable trajectories leading to safe mission execution are of prime importance. Sonar is used. In this paper, we present a trajectory planning system which is able to take into account uncertainty about the vehicle's dynamics and nonregular terrain or underwater obstacles. We consider the problem as a differential game against Nature under state constraints. By using results from viability theory and set-valued analysis, we propose a numerical scheme able to compute approximate solution to trajectory planning problems.