Waypoint planning for Autonomous Underwater Vehicles with Terrain Relative Navigation

Planning routes, defined by a series of commanded waypoints between a start and goal location, for return-to-site missions using Autonomous Underwater Vehicles with Terrain Relative Navigation is particularly challenging due to the dependence of the navigation estimate on the path flown. Waypoints must be selected to ensure the vehicle has an accurate TRN navigation estimate to arrive at the intended target site with a high probability. This probability can be estimated using Monte Carlo simulation and used to optimize the vehicle route. An Upper Confidence Bound can also be used in the optimization to decrease the number of simulations when compared to a brute force search. For the example return-to-site mission at Portuguese Ledge, the optimization significantly increases the expected probability of ending at the target site when compared the straight line route.