Time-optimal Path Planning for Sea-surface Vehicles Under the Effects of Strong Currents and Winds
A path-planning methodology that takes into account sea state fields, specifically
wind forcing, is discussed and exemplified in this thesis. This general methodology
has been explored by the Multidisciplinary Simulation, Estimation, and Assimilation
Systems group (MSEAS) at MIT, however this is the first instance of wind effects
being taken into account. Previous research explored vessels and isotropy, where
the nominal speed of the vessel is uniform in all directions. This thesis explores the
non-isotropic case, where the maximum speed of the vessel varies with direction,
such as a sailboat. Our goal in this work is to predict the time-optimal path between
a set of coordinates, taking into account flow currents and wind speeds. This thesis
reviews the literature on a modified level set method that governs the path in any
continuous flow to minimize travel time. This new level set method, pioneered by
MSEAS, evolves a front from the starting coordinate until any point on that front
reaches the destination. The vehicles optimal path is then gained by solving a
particle back tracking equation. This methodology is general and applicable to any
vehicle, ranging from underwater vessels to aircraft, as it rigorously takes into
account the advection effects due to any type of environmental flow fields such as
time-dependent currents and dynamic wind fields.