Ships must practice celestial navigation

Training for this was discontinued, but brought back in 2016:

* https://www.npr.org/2016/02/22/467210492/u-s-navy-brings-bac...

Now if only the US (and others) would get their act together and build out a backup system to GNSS. China, for example, has built out an eLoran system:

* https://rntfnd.org/2024/10/03/china-completes-national-elora...

An old USAF video explaining how the theory works (it assumes a geocentric worldview: the Earth is the centre of the universe (but it's not flat :)):

* https://www.youtube.com/watch?v=UV1V9-nnaAs

>As The American Practical Navigator (aka “Bowditch”) states, “No navigator should ever become completely dependent on electronic methods. The navigator who regularly navigates by blindly pushing buttons and reading the coordinates from ‘black boxes’ will not be prepared to use basic principles to improvise solutions in an emergency.”

I wonder if this mindset is also applied, for example, to the rest of the military. Does the Army regularly practice land navigation? I know they get at least one landnav class, but it is a perishable skill. If you don't practice, you'll soon forget about it.

I guess this could also be useful to civilians. Being able to do stuff without relying too much on electronics.

There is a lot of GPS jamming happening these days, especially in the Baltic around Kaliningrad. Makes a lot of sense.

It's really funny to watch a bunch of people contradict the US Navy when it comes to navigation at sea. No, a cell phone isn't going to work in the middle of the Pacific and no, the US Navy doesn't use Google maps. Go get on an actual boat sometime and sail out of sight of land, you lose cell signal way before you're even over the horizon.

Have there been any computer vision systems that can approximate celestial navigation using common sensors like a camera, An electronic compass, and a tilt sensor? Something like a computer vision based auto sextant. This is an idea I have thought about for a while but I have zero background in this area.

Somewhat related - https://oceangloberace.com/

A round-the-world yacht race where modern technology is not permitted. (Where 'Modern' = pretty much anything that wasn't in common use in 1970's/1980's).

A fair few of the boats that have taken part had previously run the Whitbread Race in 80's.

GPS isn't permitted outside of emergency situations, and neither are mobile phones/computers (which are sealed in a bag by race organizers for the duration of each leg).

Celestial navigation is required iirc.

Other banned items include: - Carbon Fiber - Digital Music (Only cassette tapes permitted!)

Curious how navigation at night was not possible without expensive equipment, sounds like they were relying only on starts in the morning and evening? Are the measuring something like angle of those morning/evening stars or their set/rise times with respect to the sun?

> CIC was to notify Lieutenant Commander Stanton if the Essex deviated more than 10 nautical miles (nm) from the planned track.

Pretty impressive! USS Essex would cover 10nm in about half an hour, so not much tolerance for going off-course.

So does that mean USN ships are issued with a precision mechanical chronometer for longitude, like John Harrison's original marine chronometer that won the Royal Navy's Longitude Prize?

Believe it or not, some people still use slide rules and books on trig tables for this very purpose. They use them too. No sense in having them if you aren't competent at using them.

> when celestial navigation is all but impossible without advanced, expensive equipment (i.e., a bubble sextant).

Compared to what the Navy usually steers by, how advanced and expensive are bubble sextants?

> analog navigation techniques became relevant again

I'm not sure exactly what methods were used in this navigation exercise, but if they write down numbers with finite precision at any point in the process, then the method has at least some digital component to it. Note that digital means the use of digits, not necessarily any involvement of computers or electronics.

For example, if they take a reading on a sextant, write down a number, and manually transfer it to a coordinate on paper, then that is a semi-digital process. If they take a number and then look up some kind of trigonometric table, that is definitely digital and not analog. But if the navigation process entirely consists of analog mechanical linkages and at no point any number is read out, then I would deem it 100% analog.

> using the radian rule, steering 1 degree off base course for 12 hours at a speed of 16 knots results in nearly 3.5 nm left or right of track (565 yards per hour)

This brings up a laughable feature of the US customary measurement "system", a hodge-podge of units with no coherent logic to it: 1 nautical mile ≈ 2025.371... yards, an awkward number that isn't even whole. This is because 1 nautical mile = 1852 metres exact and 1 yard = 0.9144 metre exact. Converting between these units would be a pain. (Whereas at least 1 mile = 1760 yards exact.)

Analogously, let's say you're piloting an airplane at an altitude of 15000 feet and have a horizontal distance of 7 nautical miles to your landing site. What would be the descent angle if you flew down in a straight line? The answer is not immediately obvious because you can't do trigonometry on different units. Whereas if I said you're 4572 m high and 12964 m horizontal distance away, then the angle is arctan(4572/12964) ≈ 19.4°. And even if distances are reported in kilometres, even a child knows that 1 km = 1000 m.

A smartphone has all the sensors: tilt, clock, camera. Even compass, though hardly needed. This should be enough to build an app to determine position at sea.