#GNSS #NAVISP #MachineLearning

NeuroNAV

NEURONAV is a project financed by the ESA NAVISP Element 2 program. It is a solution designed to enhance the accuracy and reliability of GNSS receivers, employing Machine Learning (ML) to reduce the impact of multipath errors.

Related Articles
Gallery
Get Yours Now

NEURONAV is a project financed by the ESA NAVISP Element 2 program. It is a solution designed to enhance the accuracy and reliability of GNSS receivers, employing Machine Learning (ML) to reduce the impact of multipath errors. We have developed both the hardware and the software required for dataset preparation, as well as the training and validation of the machine learning model. 

It focuses on maritime transport, targeting ships in transit or operating within ports where it correlates multipath errors with the geometry of the vessel to improve performance. By leveraging the spatial geometry of the environment in which the receiver is situated, the algorithm can be utilized to estimate the positioning error for multipath mitigation. 

First, we analyzed different neural network architectures and training algorithms while also creating a list of requirements for the product based on the latest maritime and industrial standards. We constructed a list of requirements and additionally, a list of recommendations was created as a guideline that could help the integration of the system. 

Then we analyzed the current hardware and software components to identify opportunities for enhancing their design. During this stage several key tasks were completed: selecting parameters from the SBF messages from the GNSS receiver to be used as inputs for the ML model, preparing datasets, and establishing the testing methodology. 

During the Test Readiness Review (TRR)  we completed the prototyping task and we began the testing campaign, when the system was deployed in a maritime environment. The campaign involved testing on two different vessels, both operated by the Maritime Hydrographic Directorate (MHD). 

Today we are getting ready for the final milestone which marks the end of the project, and this involves a presentation of the entire development process, challenges met, solutions that were implemented and our conclusions on the subject. This is expected to happen in the near future.