That is correct provided the wind data is removed. the bigger problem comes when it is given faulty data and treats it as a known error.
Now days a lot of DP2 vessels have 3 independent sensors as well. You bring up an excellent point that you need three for the system to get a more accurate account of which reference input may be in error.
The Pilots on the Norwegian coast kept their own course book, giving course and distance on every leg of the inshore route. Most of them would be certified only for certain section of the coast, but there were also a few oldtimers that was certified to pilot ships from the Swedish boarder in the south to the Russian border in the North. (This was terminated sometime in the 1960’s)
The Skippers on the small freighter operating on the coast, like this one:
would each have their own course book, with distance and time between course changes.
Their main aid to navigation was an alarm clock (or stop watch for shorter legs). They would lash the wheel and run down to grease the old Semi-diesel when on longer open stretches.
The marking used on the Norwegian coast is based on the sector system, with some lights having a dozen or more sectors:
The stop watch also came in handy to ensure you didn’t miss a narrow sector while the light was dark.
PS> My experience from Norwegian coastal sailing is limited to the years 1964-66, so i cannot claim to be an expert, but I did start my own Course book.
Sailing the coast at 36 knot on MTBs for a year honed your skills though. With open bridge and the radar in the chart room below, a lot of the navigation was on pure visual and coastal knowledge.
I don’t know anything about DP and my understanding of Kalman filters is based on reading a couple papers which I only understood the broad outlines.
Based on that, I think a Kalman filter would have a better than 50% chance because its going to take a weighted estimate of new information based on prior information. Kalman filters are based upon some kind of Bayesian method which uses a weighted technique to use new information to update so-called priors.
Here’s some random thing I just found on the Internet:
the Kalman filter can be seen as a recipe to update the distribution of the state; you start with a prior distribution, reflecting your knowledge (or a difuse prior, if you know nothing) and update that distribution each time you process a new observation.
It’s very possible I am wrong about this.
This is a good read if you enjoy reading these types of things.
Your on the right track with your analysis of weighted estimates. As you well know the more inputs the better weighted average. With just two inputs in a weighted mode your vessel could be put way off a desirered position. If you have three or more it helps eliminate that problem.
Thanks, looks interesting:
The estimation problem solved by the Kalman filter can be expressed as follows: how do
you optimally estimate the state of a vessel with an approximate knowledge of the vessel
dynamics (imperfect mathematical model) and with noisy measurements from sensors
and position reference systems? What is the best state estimate you can get out of all that?
I came across this from the use of informal Bayesian (which Kalman filters use) for problem solving.
This is an 11 minute video but 1:50 to 3:50 shows how updating priors works.
Deja vu! How did I miss those posts?
I remember the Snowbird in Seattle around 1990(?) when it was being converted for tuna fishing in the southern Pacific. A lot of the work was done at the old Marco yard in the ship canal. years later, around 2007 or 2008 I saw it again in San Diego where it was tied up at the Tuna Wharf on the embarcadero after a fire. I never heard anything about it after that and believe it may have been scrapped. It was worth a fortune in copper nickel piping alone. It had quite a history as a standby vessel for the “Texas Towers” off the East Coast as well.
I got a great tour while it was at Marco and recall that the galley had Mr. Roberts movie posters and a plaque claiming it to be the boat that was in the movie. There doesn’t seem to be any solid information to confirm which of the FS boats were used in the movie. It will probably take a while to find them but I have photos of the boat stored away in a box someplace and will scan and post them when I find them.
Thanks for posting those photos, it brings back memories of similar trips on the Coastal Sea as a relief chief. Now I look at it as dues paid.
That depends wholly on what information the DP manufacturer allows the Kalman filter to see. Most good mfgs do everything possible to prevent the Klaman filter from seeing dirty data.
I took a trip off to fly back to Maine for family business and when I got back to Seattle they told me that the Capt, not Doug, had run the Snowbird aground while going astern. Tore off the rudder, rudder went through the props and punched a few holes in the hull. Might have been 1986 or '87.
That was the end of Express Marine Transportation. Their other ship, the Sea Producer had already been seized by the federal marshals in a lawsuit. Express Marine vs Trident Seafoods.
The Westone was the FS that Mr Roberts was made on. I believe it sank in the late eighties under rather interesting circumstances on its way to the Philippines.
Does bring back memories, my time was on the old Coastal Trader, Made a few trips with with Capt Doug
when I wrote about sailing with just a magnetic compass with an oil lamp to illuminate the card at night I never associated it with this topic. Was that where the stinking compass came from?
At the risk of being like the new hand who was sent to the galley for dish soap to lubricate the steam whistle, you can find the source of the topic title by googling on “stinking badges.”
There is a whole system behind this method. The Ballard fisherman would transit to Alaska with several boats at a time. One boat, with an experienced fisherman who had made the trip many times would lead and the other boats would follow.
Capt Doug started out as deck hand/cook on one of the follow boat and worked his way up to captain of lead boat after a few seasons.
I remember the first time I crossed Wright Sound. Doug came up to make sure I got lined up on Grenville Channel OK. He told me a story that a line of boats had gone the wrong way one time and went a long way up the wrong fjord. He said it was someone else but I had the suspicion it was him running the lead boat.
When he came up I’d already gotten the course off the chart, 330 degrees or thereabouts. IIRC. After I made my course change Doug didn’t go below till we could easily see that we were in fact headed for the right opening.
Thanks for that, I had never come across the term before. Consider it filed along with asking the Chief Engineer for a “long wait” and green oil for the starboard lamp.
The old timers running between Seattle, Tacoma, and Vancouver to Prince Rupert and Alaska also used a course book. Captain Hansen’s is probably the most famous. It was used on the Union Steamships running up the inside in all weather conditions.
They used the compass, clock, and whistle and rarely had to slow down for night or fog or snow.
For the uninitiated, the whistle told them distance off the beach. I began my seafaring career working towboats running between Seattle, the panhandle, and the Aleutians. There were still old timers working on the boats who had amazing stories of running the inside passage before radar. Even when I was doing it there was still an occasional grounding, mostly because the watch fell asleep and missed a turn.
It would be lumped in with the other unknown errors that it calls “current”. The only reason that would effect station keeping is because the wind data is prioritized and bypasses the model so the thrusters can counter a gust instantly rather than waiting for the model to process the change.
With regards to using local knowledge and techniques as opposed to being set-up to cope with a large range of situations.
This is from the CG report on the grounding of the Kuluk.
A license as master or mate of towing vessels endorsed for Oceans authorizes service on oceans and on the subordinate routes of near-coastal and Great Lakes– inland waters (except Western Rivers). In general, the AIVIQs officers could have towed anything, anywhere in the world. Examples of the scope of possible towing operations would include towing a disabled bulk carrier in the South Pacific, a disabled cruise ship with passengers aboard in the waters of the Antarctic or towing an object of unique and unusual design. The issuance of the license and endorsement do not take into account the maritime environment. Towing on the world’s oceans, the licensed towing officer can encounter frigid mountainous seas, sandstorms, tsunamis, coral reefs, poorly charted areas where earthquakes have changed the bottom topography and a host of other unique operational considerations. The ocean’s license endorsement makes no practical distinction for these conditions which effect towing operations.
One point is that to get a good understanding, the entire picture has to looked at, not just certain elements.
By way of example say you have a mechanic Al, who only works on one make and model engine (Cat 3412) but knows that engine inside out. Can overhaul it blindfolded. Doesn’t need no stinking manual.
Now take another mechanic, Bud. Bud and his crew can overhaul almost any marine engine encountered, but they are slow and require a manual for each engine. They cannot work blindfolded.
Not a perfect analogy and there’s going to be some overlap.
Al and Bud cannot be compared directly, they are using different systems. It might be a big mistake to pull Al out of his shop and have him work on a different engine. However Bud might miss subtle points with regards to the Cat 3412.
Likewise, Capt Doug didn’t need a compass. But his system had limits, he refused to sail in areas he’d never been. He’d tell the office on the MF/HF they they were weak and unreadable.
By contrast a unlimited deep-sea officer should be able to sail anywhere in the world. But he needs his compass, his charts, planning guides, nav equipment - a pilot to go into port. Not to mention four years sitting in classrooms. Doug didn’t need no stinking education, didn’t need a no stinking pilot.
BTW - this applies to the Navy system as well. It’s a mistake to focus is on the number of people on the bridge as compared to a deep-sea vessel taking cargo from A to B. It’s just a single element without considering the rest of the system, including what the Navy’s requirements are.
I couldn’t imagine that I might have something to contribute to this thread, being by training a deep sea mariner, but I realised that a summer I spent on an Isle of Wight car ferry fitted pretty well. The ferry ran from Gunwharf in Portsmouth Harbour to a dredged channel at Fishborne out on the island. Once off the berth we would turn to port and run along parallel to the shore and then head down the buoyed channel to towards the Martello Tower, built to protect the town against Napoleonic invasion in the 1790s. At this time the instruction to the helmsman was “Steady she goes”. We would then tell him to set course for “The Trees” which was a hilltop on the Isle of Wight. Apparently there had once been a clump of trees on the top. Heading back, when we turned towards the harbour the instruction would be “Foudroyant” which was an old wooden warship, once the 38 gun frigate Trincomalee (built in 1817) anchored well up the harbour. During the time I worked on the ferry the Foudroyant was taken away for restoration leaving a space. We still gave the helmsman the “Foudroyant” instruction and he still headed for the space the warship had formerly occupied.