Why Integrate the Autopilot ?

[colemj]

Originally Posted by sidmon

Again...NOWHERE will you see that I advocate dispensing with modern marine electronics.

I wasn't responding to you.

Mark

[s/v Jedi]

Originally Posted by sidmon

Yeah flying is safer than it has been...If one simply looks at the macro stats which have the accident rate to the right of a decimal point and preceeded by several zeroes. And the recent media attention on that which intimates our problems are solved scares the beegeebus out of me...

Second point is the lamentable lack of weather training in both the aero and nautical worlds (note the need for weather forum thread in this venue).

Expect to hear more from me on that score.

I am sure Nick will attempt to lecture me (on things he has no clue about) and Mark will misquote me there too...

Why would I? You just started to post sensible things like agreeing flying has become safer. I think you just need to get used to people questioning your statements on public forums and supporting your statements with solid references instead of just "because I am expert and say so".

I learn new things on CF every day. Something I learned a long time ago is that it isn't really helpful to post things like you do above on me not having a clue. For all you know I was navigating the seas when you were still dreaming about it. You don't know me well enough to make these statements; the only reason you do it is because I don't agree with you. It would be more helpful for the thread if you reply to references brought up against your statements instead of jumping over them and just popping out new statements.

For example, we're still not done with he Port Royal. You state the grounding was caused by complacency. The US Navy has refused to make a statement on he cause but has retired the captain. I have stated the captain was incompetent. You need to explain why you know more than the rest of the world, with references. For now, my statement is in line with the reference I brought forward and I see nothing supporting your claim. But you just stopped there and start calling me clueless, trying to avoid further discussion on the topic you brought forward yourself.

cheers,
Nick.

[colemj]

Originally Posted by sidmon

"Integrated electronics allow stupid people to look smart until they hit something easily avoided, perhaps killing themselves or others in the process."Not sure where you got this from (went back and looked but its not in any of my posts), but I never said it.

Try to keep up here. Let me repost what I posted on #208 (with added bold emphasis to help out):

"Weellll.... This is your quote and challenge I responded to (I felt free to go back and look):

"Never. Not Once. Has anyone ever said that here Mark."

And I showed you an example of someone saying that here. And further examples of deep implications of that sentiment (integration leading to head on collisions, lost legs, etc).

Maybe your real name is "anyone"?"

Mark

[goboatingnow]

Boy this thread has descended into vitriol
Dave

[Dockhead]

Originally Posted by s/v Jedi

I have never made such a mistake. The worst I did was long before GPS and it was taking longer than I wanted to find a marker with the binoculars. This prolonged the trip because I stayed in safe deep water until I found the marker leading to a channel between sand banks.

I think that the big majority of professional mariners never made such a mistake while on duty either.

Well, you are a supermariner then, neither mortal nor human, as far as I'm concerned, rather some kind of Sailing God.

As a mere mortal sailor, I work like hell to avoid mistakes on my boat. The consequences of making a mistake on my present boat, which displaces nearly 25 tons, are much greater than the previous boat.

In thousands of miles I have never once touched the ground on my present boat, and I have worked like hell to reduce to the practical minimum any risk of that happening. That is because when you have a 5 or 6 meter range of tide, like here, and a 2.5 meter deep fin keel, the consequences of running aground can be catastrophic. That means checking, double-checking, triple-checking rise of tide calculations, tide timings, charted depths, etc., double and triple checking where I am (using visual bearings besides the chart plotter) when running a difficult entrance or channel, double and triple checking buoyage, etc., etc., etc., etc.

Nevertheless, unlike you it seems, I am mortal and human. At night once, in pitch darkness with no moon, navigating through the twisting channel leading through shoals to the Hamble River, I saw a green lighted buoy which seemed to be part of the series I was following, and failed to iniate a turn at the right moment to follow the channel. I was double-checking myself against the plotter and caught the mistake in time -- but it was a mistake. The buoy belonged to a different channel altogether -- on that moonless night you could clearly see the flash, but couldn't discern the distance, and there is a confusing mass of different flashing buoys in Southampton Water.

Oh, I could bore you with plenty more examples. All without any negative consequences, because I caught all of my mistakes in time. But -- there but for the grace of God go I. I envy you your total confidence, which I don't share. If I were as confident as you, I would surely have run aground by now, maybe with terrible consequences. I am always nervous, always checking and double checking, never relying on one source of life-critical information, never relying on any calculation until I have rerun it a couple of times, and always nervous about making a mistake, when I am piloting my vessel in dangerous waters.


As to Schettino -- yes, I agree with you, according to the information we have (which, however, might not be true or complete), it seems that he made a grandiose screwup. Recklessly approaching a known hazardous area without using any of the tools at his disposal (much less multiple ones) -- doing it "by eye", if that is what he really said. No, like you, I have never done anything like that, and can't imagine doing it, not on my 25 ton yacht, and inconceivable on a bajillion ton mega cruise ship. Still, just because we wouldn't commit that kind of gross error doesn't mean we're immune from smaller errors, which can nevertheless have catastrophic consequences.

[colemj]

Originally Posted by Dockhead

I have only recently started using "track mode", which steers the boat to keep her on the rhumb line to a set waypoint. This is useless -- in my opinion -- for sailing, or even dangerous -- because the pilot will change course without regard to sail trim, which could result in an unexpected jibe or other problems. Besides that, there are not that many situations where you actually want to stay on the rhumb line. If you are being swept by tidal currents, you're better off steering a constant heading and doing a tidal vector calculation. Nevertheless, "track mode" is useful in limited situations -- when you are motoring, and when you are trying to follow a certain path where there are obstacles on either side, making it desirable not to deviate from that path. Usually I would hand-steer in such a situation, but "track mode" is actually very helpful on a few occasions.

The best part of track mode is that it significantly calms down the autopilot in large or confused seas. In plain heading mode, confused seas can slap the boat around and the AP is always sawing back and forth making corrections to these small heading changes. In track mode, the AP works more off XTE and small, transient course changes from being slapped around are ignored. The pilot settles way down and the track through the water is straight as an arrow.

Our new autopilot has a mode called "wind-nav" which is brilliant. It steers to VMG, learns your tacking/jibe angles and calls the tack/jibe points to maintain VMG (you have to OK them manually before it performs the course change). Of course, the pilot needs to be connected to navigation and wind data sources...

Mark

[Dockhead]

Originally Posted by colemj

The best part of track mode is that it significantly calms down the autopilot in large or confused seas. In plain heading mode, confused seas can slap the boat around and the AP is always sawing back and forth making corrections to these small heading changes. In track mode, the AP works more off XTE and small, transient course changes from being slapped around are ignored. The pilot settles way down and the track through the water is straight as an arrow.

I would still never use it while sailing. If you are taken off the rhumb line by a current, the pilot will turn -- often sharply -- to take you back to the rhumb line. And won't warn you or ask your permission. If you are sailing, you will at best be way out of trim, and at worst you will destroy gear with an unintentional jibe. Track mode is not for sailing -- it's actually dangerous.

Originally Posted by colemj

Our new autopilot has a mode called "wind-nav" which is brilliant. It steers to VMG, learns your tacking/jibe angles and calls the tack/jibe points to maintain VMG (you have to OK them manually before it performs the course change). Of course, the pilot needs to be connected to navigation and wind data sources...
Mark

This is really cool -- never heard of that before. What kind of system do you have? Can you describe it more? Sounds amazing.

[colemj]

Originally Posted by Dockhead

I would still never use it while sailing. If you are taken off the rhumb line by a current, the pilot will turn -- often sharply -- to take you back to the rhumb line. And won't warn you or ask your permission. If you are sailing, you will at best be way out of trim, and at worst you will destroy gear with an unintentional jibe. Track mode is not for sailing -- it's actually dangerous.

I don't agree that it is dangerous. How would a current suddenly and sharply shift you off a rhumb line (assuming you are not passing by an outflow of a river or cut close to shore)? Any current I've ever been in has simply set me to an offset direction and the AP slowly (very slowly) compensates. Even set to track mode, it steers to a bearing and not an imaginary rhumb line, so it will never just turn sharply to bring you back to a line. If it does, you have a dangerous pilot - I have never used one that behaved in that manner.

We use track mode a lot when sailing and passage making and have never had it do anything remotely dangerous.

Originally Posted by Dockhead

This is really cool -- never heard of that before. What kind of system do you have? Can you describe it more? Sounds amazing.

The current Simrad and B&G pilots have this feature. We have the Simrad, but the B&G just has a different sticker on it. I don't know how to describe it any better - the AP takes both nav and wind data, calculates VMG based on current performance, calculates laylines, steers to VMG and calls for optimum tacks. Initially, the user sets the unit up with a proper general tacking angle for their boat and this is used until enough tacks have been made for the AP to know the performance.

Mark

[capn_billl]

Originally Posted by s/v Jedi

Sure, sure. Radars and chartplotters show false info, got that.

cheers,
Nick.

The info doesn't have to be false, to be misinterpreted, or incomplete.

And yes both show false info sometimes.

Chartplotters often have old datum, sometimes the ocean moves, hurricanes relocate debris, and wrecks. I once spent an entire day cruising 10 miles on dry land according to my chartplotter, (the channel had been moved to a new location).

The earth is not perfectly round, this will sometimes put island shoals several hundred feet to several miles from actual location depending on datum set used.

Now for RADAR, power and discrimination settings, can show false targets, or miss objects of low reflectivity. There are several deadspots in any RADAR caused by limitations of antenna design, or installation. A perfectly good target will simply either not show up, or show up off of it's actual position due to reflection from a sidelobe. Atmospheric lensing can also cause targets to move, or be hidden, (same effect as mirage, or heat ripples on hot pavement). Lightening, (or invisible ionic charges in the air), creates bursts of microwaves that can fool a radar, (or ping it's gain settings).

I do integrate my electronics, but I also have a second independant handheld GPS to verify. A bad comm cable can wreak havoc with position data.

A too common mistake is to run into a clearly visible object while looking head down on the many displays.

[St. Elsewhere]

Originally Posted by sidmon

Rush-Hour Read: How Automation Leads to More Aviation Accidents

Pilots’ “automation addiction” has eroded their flying skills to the point that they sometimes don’t know how to recover from stalls and other mid-flight problems, say pilots and safety officials. The weakened skills have contributed to hundreds of deaths in airline crashes in the last five years.

There is a difference between pilots who were type rated prior to today's auto flight systems, and the younger ones. They spend far less time hand flying, and do not seem to recognize and correct for manageable equipment failures as proficiently. Also, even experienced pilots have an expectation for the systems to work.

The AF A340 is an example.

Also the Turkish 737 at Schiphol (Amsterdam) is another. A young Fist Officer was conducting an auto-land approach when a Radio Altimeter failed about a mile from the runway. (A radio altimeter is like a depth sounder - it measures the height above ground). The failure caused the auto-land system to retard the thrust. The failed radio altimeter was telling the system that they were close to ground and time to cut power, and land.

The First Officer, and in fact two other pilots who were in the cockpit should have recognized that they were not ready to cut power. The First Officer should have simply advanced the thrust levers, but did not. All three should have been focused on airspeed, altitude, and runway alignment throughout the approach.

The Captain took the controls too late, and they crashed short of the runway.

Despite these problems aviation is more safe with it than without it. And for recreational boating the same is even more true. There is far less risk of depending on it. You can't stall a boat and drop to the bottom of the sea.

[goboatingnow]

I would still never use it while sailing. If you are taken off the rhumb line by a current, the pilot will turn -- often sharply -- to take you back to the rhumb line. And won't warn you or ask your permission. If you are sailing, you will at best be way out of trim, and at worst you will destroy gear with an unintentional jibe. Track mode is not for sailing -- it's actually dangerous.

I know of no pilot thats acts like that without asking for a confirm course change. There is an issue with certain pilot software that ON confirmation, it executes an abrupt change where XTE errors are gross. This usually happens where the sutopilot was taken manually out of track mode and the boat deviated signifcantly from the programmed course, then the operator attempts to re-acquire the course , with a now large XTE.

But this does seem to be changing, I notice on the latest Raymarine, you get asked whether to return to the old course or accept that there is now a new course to the waypoint.

[goboatingnow]

Originally Posted by St. Elsewhere

There is a difference between pilots who were type rated prior to today's auto flight systems, and the younger ones. They spend far less time hand flying, and do not seem to recognize and correct for manageable equipment failures as proficiently. Also, even experienced pilots have an expectation for the systems to work.

The AF A340 is an example.

Also the Turkish 737 at Schiphol (Amsterdam) is another. A young Fist Officer was conducting an auto-land approach when a Radio Altimeter failed about a mile from the runway. (A radio altimeter is like a depth sounder - it measures the height above ground). The failure caused the auto-land system to retard the thrust. The failed radio altimeter was telling the system that they were close to ground and time to cut power, and land.

The First Officer, and in fact two other pilots who were in the cockpit should have recognized that they were not ready to cut power. The First Officer should have simply advanced the thrust levers, but did not. All three should have been focused on airspeed, altitude, and runway alignment throughout the approach.

The Captain took the controls too late, and they crashed short of the runway.

Despite these problems aviation is more safe with it than without it. And for recreational boating the same is even more true. There is far less risk of depending on it. You can't stall a boat and drop to the bottom of the sea.

There is a fundamental difference, modern avionics flies the plane, the pilots are button pushers. Especially in failure modes, pilots do not understand the control logic, relying on slow checklists, also often crucial intermediate levek data ( Angle of Attack, being one) is often not displayed. Pilots are then faced with a either to intervene ( and balls it up, aka AF447) or not intervene(and balls it up, Tirkish 737)

We are nowhere at that level of automated control nor will be ever be as sailing is essentially a manual process.We are not bomarded by vast amounts of necessary sometime conflicting data. Its an apples and oranges situation

Aircraft errors such as these are often due to the incredibly complex nature of the control system and its less then adequate human controller ( and inadequate HMI). Where you have a full control system, this generates problems especially if you have to build a system that is both automatic and manual/automatic. Then add reaction times to disaster that are measured in a few minutes and its a very difficult problem to solve.

[nigel1]

Originally Posted by colemj

Our new autopilot has a mode called "wind-nav" which is brilliant. It steers to VMG, learns your tacking/jibe angles and calls the tack/jibe points to maintain VMG (you have to OK them manually before it performs the course change). Of course, the pilot needs to be connected to navigation and wind data sources...

Mark

That does sound neat Mark.

You've just prompted me to go read about VMG wind. Its indicated on my chart plotter, but up until today, I have not paid it much attention. Now I shall.

[goboatingnow]

Ill give a more reasoned approach as to why I never use wind vane mode. ( and before we start yes I know all about it and have tried it).

Firslty in a cruising boat who really cares about VMG. I care about a safe route first, then the route that makes the boat comfortable.Only then will I consider the most efficient direction

Secondly VMG is only useful where you are constrained by not being able to sail free. I try and avoid that and if is really bad, Ill probably motor. ( teh gentlemen doent sail to windward rule!)

Thirdly , if you sail by wind and dont set the alarms properly , you have have significant course shifts and for the complacent , that gets you into trouble, With bearing waypoint that I use, I will hear the sails stall , and I can retrim, but at least I dont deviate from my safe route inadvertently.

As a protection against gybing etc, that's a fallacy, most accidental gybes are caused by waves throwing the boat through the wind and because the boat is being sailed too deep for the circumstances.Autopilots no matter what mode will not save you

And finally , all wind steered autopilots I know have a wind shift course change limit error, when such an error occurs the autopilot drops out of control ( very very inconvient when youre in the head) The boat effectively is out of control at that point. AT least with track mode it takes a huge boat yaw to effect that.

From my sailing experience, wind shifts are slow gradual and anticipated, it some situation such as under thunderclouds I would consider wind directed steering as very very inappropriate.

Nop its not for me. ( and funnily I know very few, who use it).

Dave

[colemj]

Wow, the dangers our autopilots expose us to! I'm surprised they are even legal...

Mark