Open jweisbaum opened 3 years ago
This is a great idea and a priority for me. I can't make any guarantees but I've researched the problem some and look forward to the chance to implement it. :)
For context: my non profit has global nautical charts for the entire world, and we run sailing races that do real time geospatial analysis on the fly. With H3 or s2, we're interested in running shortest path algorithms in real time that take into account coastlines, nautical charts and weather. This could totally change how sail drones and major ocean races operate and improve the safety of all offshore navigation.
What are H3 and S2?
They are both libraries for creating spatial indices. S2 is by google and converts every spot on earth to a location on a Hilbert curve. H3 splits the world into hexagons with unique ids. https://s2geometry.io/ https://eng.uber.com/h3/ https://h3geo.org/
Both libraries have open source C/C++ implementations.
This issue has been labeled inactive-90d
due to no recent activity in the past 90 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.
This issue has been labeled inactive-30d
due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d
if there is no activity in the next 60 days.
Is your feature request related to a problem? Please describe. Yes. We want to use H3 and S2 for pathfinding. We'd like to break the Earth up into tiny cells near the coasts, and larger cells in the oceans, and then use cuGraph to route sailboats based on the wind.
Describe the solution you'd like A clear and concise description of what you want to happen. We'd like to build H3 or S2 indices for huge amounts of data in parallel.
Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.
Additional context Add any other context, code examples, or references to existing implementations about the feature request here.