PREParE SHIPS - PREdicted Positioning based on Egnss for SHIPS
Background & Objectives
"The objective of the project ‘PREParE SHIPS’ is the development and demonstration of a collaborative resilience navigation solution. It aims to develop and enhance existing software solution by exploiting the distinguished features of Galileo signals as well as combining it with other nautical information on internal as well as external parameters and sensor technologies. The final navigation decision support tool implemented consists of collaborative exchange by ship2ship communication of dynamically predicted future position based on resilient position from Galileo receivers. This will increase safety and efficiency significantly and will be the base of future autonomous operations.
Besides the use of vessel on board sensors, ‘PREParE SHIPS’ will also make use of data learning of earlier ship behaviour to exchange near-future positions with vessels in the vicinity and VTS centers (Vessel Traffic Services) to increase safety and improve decision making. In order to define the correct requirements for the PREParE SHIPS combined positioning solution, a collaborative automated vessel application will be defined and developed. The vessel application will rely on the high availability positioning solution and use it to couple its various navigational systems with ship2ship/ ship2shore and aggregate information received from other connected vessels. As there will be a transition period where a lot of vessels are neither connected nor automated, solutions having high impact during low penetration are in focus. ‘PREParE SHIPS’ will implement and demonstrate a fairway geo-fencing with high precision positioning taking into account various data sources (e.g. wind and current) as well as a traffic monitoring and predicted positions so it can allow for safe decisions based on robust data. This means that ‘PREParE SHIPS’ also will implement perception layer sensor fusion that uses information collected historically in similar conditions based on machine learning-hybrid models."