Sailboat routing consists in computing the best route for a sailboat taking into account the characteristics of the ship and environmental data such as weather forecast. In the context of sailing races, the best route computation is usually based on the isochrone algorithm, a sub-optimal solution to optimize the time to destination (TtD) criterion by computing a route as a sequence of waypoints. In this paper, we propose to compute a set of possible routes by considering two criteria: the time to destination and the stress. The time to destination is evaluated according to weather forecast and boat polar diagrams. The stress function is a combination of human and environmental factors. The set of possible routes are then obtained by using an iterative multiple objective optimization algorithm. Isochrone algorithm is used for initializing the set of routes. Then mutation operators are used to explore alternative solutions. Applied to realistic test cases, our search strategy allows to obtain routes with very different characteristics in terms of time to destination and stress values, asserted by experimented sailors. Concerning the main objective of minimizing time to destination, we are competitive with commercial software such as MaxSea or Adrena.

Keywords:
stress function, evolutionary algorithm, sailboat, algorithm, operator, waypoint, routing, machine learning, boat, weather forecast
Subjects:
Information Management and Systems, Artificial intelligence
This content is only available via PDF.
2019. The Society of Naval Architects and Marine Engineers
You can access this article if you purchase or spend a download.