Depth prediction for tidal data using Harmonic Analysis method vs Neural network for tidal data
Budget: £20 – £250 GBP
Project Description:
This is full description of project
WAIT16 - Depth prediction for autonomous navigation
Water depth information is important for navigation. The level of the
sea can vary by several meters and the state of the tide can be the
difference between safely entering a harbour and running aground. With
an eventual view to autonomous navigation and passage planning, we
would like to have programmatic access to this information.
Free software tide calendars exist, in particular Xtide [1]. Xtide,
however, is designed for human use, with a graphical user interface
and a command-line interface. It does include a C library for
accessing its database of stations (location) and harmonic constants
and a C++ library for predicting the tides at given stations.
This project is conceptually simple: given a position and a time,
compute the expected depth. This requires knowing both the height of
the tide and the chart depth at that position. Do this in a way that
makes this question convenient to ask, and is as lightweight as
possible. Demonstrate this capability from a high-level language such
as Guile/Scheme or Standard ML or Go or even Python. Lightweight
means Java, JVM languages and their ilk are out of the question.
This project requires a good level of familiarity with UNIX/Linux
development, a good level of comfort with command-line environment,
and literacy in C, C++ and a suitable high-level language. Familiarity
with geographical data formats and libraries would be an
advantage. Some domain-knowledge about navigation would be helpful but
not strictly necessary.
[1] https://flaterco.com/xtide/
I have requirement to compute depth prediction using neural network model.
To make it simpler, we will be taking data of three points and predicting data for the fourth point
This is full description of project
WAIT16 - Depth prediction for autonomous navigation
Water depth information is important for navigation. The level of the
sea can vary by several meters and the state of the tide can be the
difference between safely entering a harbour and running aground. With
an eventual view to autonomous navigation and passage planning, we
would like to have programmatic access to this information.
Free software tide calendars exist, in particular Xtide [1]. Xtide,
however, is designed for human use, with a graphical user interface
and a command-line interface. It does include a C library for
accessing its database of stations (location) and harmonic constants
and a C++ library for predicting the tides at given stations.
This project is conceptually simple: given a position and a time,
compute the expected depth. This requires knowing both the height of
the tide and the chart depth at that position. Do this in a way that
makes this question convenient to ask, and is as lightweight as
possible. Demonstrate this capability from a high-level language such
as Guile/Scheme or Standard ML or Go or even Python. Lightweight
means Java, JVM languages and their ilk are out of the question.
This project requires a good level of familiarity with UNIX/Linux
development, a good level of comfort with command-line environment,
and literacy in C, C++ and a suitable high-level language. Familiarity
with geographical data formats and libraries would be an
advantage. Some domain-knowledge about navigation would be helpful but
not strictly necessary.
[1] https://flaterco.com/xtide/
I have requirement to compute depth prediction using neural network model.
To make it simpler, we will be taking data of three points and predicting data for the fourth point