Losing the Plot - Navigation / Piloting / Collision Avoidance

An interesting thread. I note that no one has mentioned the grounding of ROYAL MAJESTY off of Nantucket in June of 1995.

Ref: https://towmasters.files.wordpress.com/2009/03/ntsb_royalmajesty_grounding.pdf

“The National Transportation Safety Board determines that the probable cause of the grounding of the Royal Majesty was the watch officers’ overreliance on the automated features of the integrated bridge system…”

“After the accident, representatives from Majesty Cruise Line and the Coast Guard examined the Royal Majesty’s GPS antenna and receiver. They found that the GPS antenna cable had separated from the factory connection at the antenna. The antenna cable, which was factory assembled, showed no sign of physical damage, other than having been separated from the connection.”

Another source mentions that on the previous passage –
"Satisfied with the accuracy of the fix he just laid on the map, the navigator walked away and joined the captain, who was standing by the large windows. Above them, on the flying bridge, the loose GPS cable was banging freely in the wind and slapping the flying bridge’s deck; the connection between the GPS cable and the antenna was coming apart. At 2:52 P.M., interruptions in the global positioning data began: after a few seconds of no position signal, a one second alarm chirp sounded, similar to a wristwatch alarm. Nobody heard it; it was barely audible. At 1 P.M., the captain commanded ‘‘change course to 336, take us up to Boston.’’ Nine minutes later, the global positioning system signal was back again. But it did not last long, because at 1:10 P.M., the connection between the cable and the antenna broke away for good. "

Related - aside. Many year ago I had a safety job at a New Jersey refinery. One ship that was destined to go to the refinery with our oil on board was reported aground in Stapleton anchorage. I was discharged to go out and observe only since our oil was aboard. As i rode out on the launch the ship was broad to the current and the chain was under significant strain.

I made it to the bridge and introduced myself to the Master, and advised why I was there. At which point he kept pointing to the electronic chart and insisting he was not aground. At which point I asked him why he wasn’t pointing the same way as every other ship in the anchorage.

Reliance on electronics can make you disbelieve your own eyes. Especially if they are giving you the answer you want.

1 Like

I mischaracterized this as instrument vs non-instrument which is not the best way to think about it.

This is also from the linked post in the OP

I usually set the trails to automatic (all targets, including land, are trailed) and relative, with an appropriate and short time interval rarely exceeding 3 minutes. That is all one has to do, once every watch. Henceforth, anything the radar picks up will be trailed. Trails which are parallel or angling away from you can be ignored. Trails which are crossing your path may be a cause for concern. Plot or watch only those on the ARPA.

Using target trails is a way of filtering. It reduces the number of targets that need to be acquired with ARPA. It works the same way as watching the relative movement visually does.

Also mentioned in the post are Parallel indexing. Both target trails and PI reduce cognitive work load.

One mistake less experienced officers make is too long a range. A related mistake is trying to think too many moves ahead. Solve the problems one at a time. If changing course for a closer vessel is going to create a problem later for another vessel, change course and deal with the later problem later.

4 Likes

Kelvin Hughes made a display for collision avoidance that refreshed itself every 3 or 6 minutes, may have been more. The display was about a 30” cathode tube that didn’t need a hood in daylight unlike radars of the period. All it showed were the tails of targets and in a vessel doing 27 knots in the English Channel we thought it was the greatest thing since sliced bread. We called it the Tadpole bowl.

2 Likes

How information is displayed is important.

With the target trails the one big ship at speed and crossing can be picked out from a screen full of squid boats that are drifting with just a glance.

In restricted waters a quick look at the difference between CMG and heading vectors on the radar is far quicker and more effective than constant monitoring of the ECDIS and leaves more time for lookout.

This is an important point. A competent mate is able to trust visual information in part because they know how to quickly extract the information required from the instruments (radar, ECDIS and others) to verify what they are seeing.

For example with target trails enabled the signal of tanker Sola TS on radar would have almost certainly been seen in the Helge Ingstad incident.

What if they can’t see?

You propose that they should spend most of their time ‘keeping a lookout’ with only occasional glances at ECDIS/ARPA but that only works under conditions of good visibility.

Comparatively putting the focus on skilled use of instruments and confirming that they’re operating correctly visually works in all visibility conditions, as well as traffic conditions where there are too many targets to effectively monitor visually.

In that case of reduced visibly they would be required to call the master. A new third mate would not be expected to be able to deal with traffic / restricted waters in reduced visibility without assistance.

It works the other way around in South China Sea levels of traffic.

I don’t know off hand how many targets an ARPA can track simultaneously but it’s far more than the 5-7 items that a person can track in short-term memory. That’s why it’s necessary to be able to filter out vessels that are not a threat but still be able track the relatively few ones that are a threat.

Here is:

A 260-strong Chinese fishing fleet lighting up the horizon

image

If those boats were ahead of the ship I’d change course until there was a clear path ahead of the ship and then steer that course. Almost all of them would be drifting, but they will be a few moving. Plus there may be a big ship or two hiding in there somewhere.

Mate is going to be busy acquiring them all and maybe too many for the ARPA to process.

1 Like

on our steady run there was always a large fishing fleet of the coast of florida - north of the bahamas. It was usually a call to the Master a few hours early to ask if he wanted to go to work that night, or if he wanted me to " go to sea " and leave them all inside and take a few hours of diversion.

1 Like

Yes, sometimes they can’t be avoided. They are not as thick as they look in the photo as they are spread out in two dimensions.

I often let the mate weave though the boats while I watch further out. That way I can keep an eye on both. The mate can keep a constant visual watch while I shift between visual and AIS/trials etc.

If the captain does it the other way around, keep the conn while the mate watches the AIS/radar the mate sometimes has a difficult time doing both. Just maneuvering through on the other hand is basically mostly intuitive.

We were pretty lucky to have a steady crew, all with lots of experience on the ship and the run. In general the mates would almost always keep the conn. The Captain was the extra pair of hands and eyes just to help handle the increased demands. And of course the ultimate safety check. Most of the time, both of our Masters kept to their chair and their radar. These are great teaching moments to give junior mates and experience and confidence.

2 Likes

The “Safety Fairways” In GOM were loaded with shrimp boats that weren’t following the LOTFW concept. Agree with a poster, I sat back but observant to help the mate figure out the problems.

Being proficient in collecting information from the various sources, so that due emphasis can be placed on processing rather than collecting it, is one important component of dealing with information overload. Visual information, requiring more experience to systematically collect and interpret, tends to take the backseat. That is the premise of this thread as I understand it.

Equally important, or perhaps more so, is understanding which sources provide the most relevant information, or in other words what the various tools are good at. The radar is the tool for telling the precise distance to various objects, the ECDIS is peerless for establishing your position and velocity vector in relation to terrain and bathimetry, with AIS overlay it is really good at visualizing static traffic situations, etc etc. This is very basic stuff, and understood on the intellectual level by all and anyone who stands watch, but understanding it on a practical level so that it becomes part of the instinctive process is a different matter.

Looking out the window is the best tool for monitoring dynamic situations, in some cases the only viable one; There is nothing quite like it for observing a fast craft running in circles, or coordinating a turn as two reference points are brought into line. I think that fully appreciating this is the key to gaining proficiency, as deliberate practice will follow once the utility is realized. As far as correlating visual information with other, relevant sources, consider the following example:

Looking out the window, we have two ships on our Starboard bow, both of them on increasing apparent bearings. On the surface this is an entirely uncomplicated situation and nothing to worry about, but a glance at the radar shows a different picture:

Now that we know how far away they are, we can appreciate that the two ships may be on a collision course with each other, and that B will probably alter to Starboard in order to pass astern of A, possibly bringing it on a collision course with us. We now have a dynamic situation in need of continuous monitoring. Simply being proactive and altering to starboard is not a solution, because it will bring us dangerously close to B’s current vector. This is where your green mate uses the ARPA list as his primary tool for monitoring the situation, setting him up for a stressful time as B’s CPA starts decreasing.

The better solution would be to monitor B’s visual aspect, and pay close attention to the rate of change of apparent bearing when the aspect starts increasing (“you see it starting to turn”), then use ARPA and what have you to verify your understanding of the situation. If your mate understands the value of his tools, in addition to being proficient in their use, he is more likely to choose such a course of action.

1 Like

That example is a good one depending upon how difficult it actually is.

One issue with using the ARPA / Radar here is when “B” first begins to turn that information will be seen very quickly visually but ARPA will update "B"s vector slowly and gradually.

The visual solution would be to wait until B makes it’s course change then steer for A’s stern showing an intention to pass astern of A. After B passes astern of A then own ship would be showing B a red light.

I think that the key to improvement in skill is to focus on deliberate practice on solving problems visually but it is very difficult to convince a green mate of that.

For the mate that has acquired the visual skills it seems as natural as avoiding collisions when walking down a crowded sidewalk or driving a car in traffic.

EDIT: This problem can be solved using the “solve the first problem first” principle but it might be a two step process. First make a course change to starboard that is “readily apparent” to B and then make a smaller alteration to port to pass astern of “A”. Better however to make it just one step such that it is also always “readily apparent” the intention to pass astern of A.

2 Likes

I think the problem get a bit harder if you add a navigation component as well. Assume there is a shoal 1 mile astern of B, and you are piloting.

In such an instance, at least in my long ago memory, that is where the balance of radar and window become important. Parallel indexing to avoid the shoal ( if possible) while addressing the collision avoidance.

Question for the more modern sailors - is that a bit obsolete in the world of instant positions on electronic charts ??

1 Like

I wouldn’t say it’s obsolete - anyone that has been to sea for awhile has had a monitor fail or a mouse fail or any number of things happen. The better solution to that is keeping your electronics reliable and having instant backups though. There are better techniques, settings, technology available now than some I hear described ITT.

I still prefer to do things visually, but that’s because I was taught by a guy that did the Inside Passage on Cap Hansen’s logbook and GPS be damned. I believe people that say that visual piloting outside of regular confirmation of electronics function and accuracy is better are reaching to come to that conclusion.

Nobody is saying that.

As captain I watch the watch officer (and pilots).

My observation is that the officer that are using the ARPA / Radar in heavy traffic for situational awareness does not have time to switch and observe bearing change of any one vessel visually and can’t do so without “losing the plot”.

It’s not a symmetrical situation wrt to time and cognitive workload. Need to use visual to first filter information to reduce workload. Doesn’t work the other way round.

Use visual to maintain situational awareness and use ARPA only on the few vessels that have been determined to be a threat.

I disagree.

You actually used an example in your pic of a squid fleet that proves my point rather well so I’ll use it too. In that fleet, imagine one vessel has lost it’s generator and is leaving the grounds for repairs. No generator, no squid lights.

You absolutely will not pick that vessel up visually. The background lights will drown it out completely. You will only identify and be able to avoid that vessel with radar.

If your instruments are not reducing your workload over your eyes you need to develop skill in utilizing your instruments to their full potential. The instruments are indisputably better than your eyes and can nowadays calculate, monitor and guard better than a human can if the human knows how to use them. All of your examples leave the ship open to a collision from a visually missed vessel and nobody has eyes that good or in the back of their head.

If they’re losing the plot they don’t know how to match up what their instruments are telling them with their eyes see. That’s a learned skill.

That’s why target trails are used.

image