Weather conditions are an important factor in every means of transport, especially in marine journeys. That is why I wanted to share information about a paper that I co-authored, at the University of Pisa, which solves the problem of weather forecasts in maritime applications.

The main goal of this paper is to propose a machine learning algorithm, which will be coded into a microcontroller and will be able to predict in short-term the wind speed weather conditions on board of a boat.

Boats and weather conditions

Maritime journeys significantly depend on weather conditions, and so meteorology has ever had a key role in maritime businesses. Nowadays, the new era of innovative machine learning approaches along with the availability of a wide range of sensors and microcontrollers creates increasing perspectives for providing onboard reliable short-range forecasting of main meteorological variables.

The algorithm was codded following specific conditions, through the Zerynth platform. Furthermore, this proposed method and algorithm have been tested on real weather data recorded both during a ship journey and by static weather stations.

This study is supported by the “Lincoln Project” (Lean Innovative Connected Vessels) – a project that was previously talked about on the blog.

Technologies and tools

The whole system is composed of:

  • Weather Station: which carries out continuous measurements on a ship.
  • Data acquisition and elaboration unit: that automatically records all the measurements of its sensors.
  • Machine learning algorithm: embedded on hardware that provides weather predictions.

weather station hardware

How did we implement this?

Onboard of the tugboat we have installed a weather station (Airmar220WX, Airmar Technology) which, through an integrated GPS and compass system, calculates the speed and direction of both real and apparent wind, calculated based on the speed and direction of the boat.

Then, this data obtained from the GPS module is indispensable for obtaining the movements and the global positioning of the vessel, and for correct data and events chrono-referencing.

See the system in action

Livorno wind speed predictions for 30 min

As you can see, the results are really impressive. Go take a look at the full paper to learn more.

If you are interested in knowing more about this research you can also contact the authors asking for a pre-print copy of the paper.

Share This Story, Choose Your Platform!

About the Author: Daniele Mazzei

Daniele Mazzei
Daniele is the CPO and co-founder of Zerynth. His strong interest in the interaction between people and intelligent objects led him to co-found Zerynth and to design connected devices and Industrial IoT applications. After earning a PhD in Bioengineering and Biomedical Engineering, he is now an Associate Professor at the Department of Computer Science at the University of Pisa.

Follow Zerynth on

Latest Posts