Underwater Navigation Device

[emoreira]

In the other thread starting at post #35 on 10/18/08, user kacem78 suggested an inertial guidance device he was working on. The discussion is technically interesting and goes on for awhile, I expect you'll find it worth reading. Alas, kacem78 last posted there in June 2009 (might be his last post, and shortly before the Navimate announcement) and hasn't been on SB since August 2010.

From the discussion, I believe kacem78 had something working as a prototype. Five years on, I expect that the state of commercially available MEMS devices makes this even more feasible. So you're quite right to suggest it, but I thought I'd connect everyone reading it with the earlier discussion of solving this with inertial guidance.

I read the whole thread where kacem78 said he was designing a device working with the Inertial Navigation concept. This was a stand-alone device mounted in the diver's back or tank that tracked the divers position relative to the starting point, without any transceiver or beacon mounted in the boat.
Since June 22nd 2009 there has been no update and no device was presented.
The idea was pretty clever and seem to be feasible.
Sorry that there was no new.

---------- Post added January 14th, 2014 at 12:40 PM ----------

More on this Inertial Underwater navigation system :
Inertial Navigation for Divers | 85th European Study Group with Industry (ESGI) - UEA 2012

---------- Post added January 14th, 2014 at 03:06 PM ----------

It seems that there is already a kit completely assembled that has also GPS interface from Honeywell : Electronic Dead Reckoning Module | Inertial Navigation System
There is also a video explaining what it does.
The DRM4000L module costs $ 550

 

[reefduffer]

...
More on this Inertial Underwater navigation system :
Inertial Navigation for Divers | 85th European Study Group with Industry (ESGI) - UEA 2012
...
It seems that there is already a kit completely assembled that has also GPS interface from Honeywell : Electronic Dead Reckoning Module | Inertial Navigation System
There is also a video explaining what it does.
The DRM4000L module costs $ 550

Eduardo,

Great finds, and excellent reading. The first study, which was sponsored/led by VR technologies, who make productised rebreather technology, makes me confident that if this navigation approach becomes a feasible technology, that there's somebody in the scuba market who both sees the potential of a product and knows how to make it happen. And the report and references are a really great resource for anyone contemplating designing something like this.

However, my interpretation of the study results leave me a little skeptical that we're there yet. It deserves a more thorough reading than I've had time for, but I've been scanning it for conclusions wrt feasibility. There wasn't much of a "conclusions" section.

From the study report, page 17-18:
A.2 Three–dimensional case
...
1. (A.2.3) With ε = 0.001ms−2Hz−1/2 as above, the values of μrˆ after one and two hours are roughly 200 m and 560 m respectively.
...

This study uses some additional sensor technology, such as electronic compass and gyros, and some rather sophisticated signal processing, but I'm struck by how close this is to my own quick swag of this several years ago. In the other thread http://www.scubaboard.com/forums/basic-scuba-discussions/209712-underwater-navigation-device-2.html, in my initial response to Kacem's post in post #82, Dec 22 2008, I said:

Kacem,
...I had spent a couple of days looking at the feasibility of building one of these from a 6-axis mems a year or so ago, I've mentioned it here on scubaboard in response to some threads on "what should we invent for scuba".

My conclusion was that the drift error using available devices I was looking at made it impractical. I think I came up with something like 1/4 mile max error for a 1-hour dive.

...

200m is ~1/8 of a mile. Still not enough to get you "back under the boat", but closer. The study was also limiting itself to inexpensive sensors like you'd find in a smartphone. Maybe that will improve, and/or a scuba product could use something a little pricier with a little less error. But the problem is that the state of the art for inexpensive accelerometer/gyro sensor tech is pretty darn good for most uses, so there's less pressure to get more accuracy in consumer sensor tech. I think there's another conceptual breakthrough needed, or some other application like gaming input that drives more accurate sensors, or this will be a long time coming to general scuba.

Which raises the question as to what accuracy would be acceptable. 20m after an hour I think is a no-brainer winner product. Kacem (post #88) said "so far i'm in the range of 1-meter drift after 50mn DIve" which I thought was astounding, and given this study you found, also a bit unlikely. If it were accurate to 50m for a 1-hour dive, I'm not sure I'd pay several hundred dollars for that, but maybe. The threshold of usefulness seems about there to me, at least at first approach.

As for the Honeywell Dead Reckoning Device - it seems to be essentially a very smart pedometer using GPS for step calibration, I think. There's a brief mention in the video presentation from the study. They brainstormed the idea of sensing fin strokes to do something similar, but I don't think that translates well.

But thanks for the pointers to these. At least get people thinking about the problem, with the benefit of other's thoughts to build on. Maybe somebody can make this work.

 

[WhiteSands]

What are your feelings about an underwater navigational device?

I think its about time.

What do you find most difficult about underwater navigation?

Lack of landmarks for triangulation. Can't figure out where you are. Maps not commonly available to divers, or if they are, are not cost efficient to buy. E.g. vacation dive.

Navigating straight lines are fine, or even round a reef/island. It's navigating from one object to another that is challenging, especially if you cannot find the first object, or if there are no maps available.

Nobody I know counts fin kicks anyway. So it's hard to measure distance. Most people use time, which is inaccurate.

How long have you been diving, and how has your answer to the last question changed as you’ve become more experienced?

2 years so far. Answer hasn't really changed.

When looking for a specific feature at a dive site (cave, wreck, etc…) how often do you have difficulty find it? If you’ve had difficulties with this, would your rather have a map to look at, or a simple arrow telling you the distance, direction, and depth to the feature?

With good directions to a single point, it's not difficult to find. I'd rather have a map and my current location on it.

Have you ever used any devices (other than a compass) for underwater navigation, and if so, what were your experiences with them?

No

How often have you been separated from your dive partner? Of those times, how many times have you easily found them, and how many took longer than you would have liked?

Only on low vis dives (< 1m vis) and one time on a night dive with many divers in multiple groups. On low vis dive, did not manage to find. Found my buddy on the night dive after about 1min+ of searching.

What would be your biggest reason for purchasing a navigation device (safety, better dives, comfort, etc…)?

Independence. I don't like following guides.

Do you use a dive computer? If so, what type do you use, how much did it cost, and what made you purchase your particular model?

Yes. $300+. Has all the features I need, looks nice, affordable and proven to be durable.

How often do you replace your computer, and what will you be looking for in your next one that your current model lacks?

Never so far. The ability to plan for Trimix, or I might just get a bottom timer.

If your next purchase of computer had an option for a practical navigational aid, how strongly would you consider the following features? Please rate from 1= Unnecessary to 5=Important, and add any comments you feel are necessary.

5. Provided it allows me to locate my buddy also in case we get seperated.

Distance and direction to boat (or dive starting location):

4

Distance and direction to dive partner (or other divers):

5

Waypoints for dive features (caves, wrecks, etc…):

5

Actual map of dive site:

5

A recorded path of your dive (including air usage and swim speed):

1

Btw, just wondering if this technology will work inside of caves & wrecks?

 

[emoreira]

Btw, just wondering if this technology will work inside of caves & wrecks?

Hi WhiteSands, the idea of Inertial Navigation Systems is to track the movements of the device related to a start point, without any input from external devices. The INS use also gravity and Earth magnetic field. In the case of caves, both gravity and Earth magnetic field are unchanged. In the case of wrecks, Earth magnetic field is seriously affected, but, AFAIK caves are dived following a life line and wrecks are dived following the ship silhouette and, if penetrated, following a life line. Caves and wrecks are not dive sites where a navigational aid is required, unless you are exploring a new cave/wreck and want to track positions and make a map.
I'm not a cave diver so, take my words as what I've seen here in SB.

---------- Post added January 15th, 2014 at 11:59 AM ----------

Eduardo,

Great finds, and excellent reading. The first study, which was sponsored/led by VR technologies, who make productised rebreather technology, makes me confident that if this navigation approach becomes a feasible technology, that there's somebody in the scuba market who both sees the potential of a product and knows how to make it happen. And the report and references are a really great resource for anyone contemplating designing something like this.

However, my interpretation of the study results leave me a little skeptical that we're there yet. It deserves a more thorough reading than I've had time for, but I've been scanning it for conclusions wrt feasibility. There wasn't much of a "conclusions" section.

From the study report, page 17-18:
A.2 Three&#8211;dimensional case
...
1. (A.2.3) With &#949; = 0.001ms&#8722;2Hz&#8722;1/2 as above, the values of &#956;r&#710; after one and two hours are roughly 200 m and 560 m respectively.
...

This study uses some additional sensor technology, such as electronic compass and gyros, and some rather sophisticated signal processing, but I'm struck by how close this is to my own quick swag of this several years ago. In the other thread http://www.scubaboard.com/forums/basic-scuba-discussions/209712-underwater-navigation-device-2.html, in my initial response to Kacem's post in post #82, Dec 22 2008, I said:

200m is ~1/8 of a mile. Still not enough to get you "back under the boat", but closer. The study was also limiting itself to inexpensive sensors like you'd find in a smartphone. Maybe that will improve, and/or a scuba product could use something a little pricier with a little less error. But the problem is that the state of the art for inexpensive accelerometer/gyro sensor tech is pretty darn good for most uses, so there's less pressure to get more accuracy in consumer sensor tech. I think there's another conceptual breakthrough needed, or some other application like gaming input that drives more accurate sensors, or this will be a long time coming to general scuba.

Which raises the question as to what accuracy would be acceptable. 20m after an hour I think is a no-brainer winner product. Kacem (post #88) said "so far i'm in the range of 1-meter drift after 50mn DIve" which I thought was astounding, and given this study you found, also a bit unlikely. If it were accurate to 50m for a 1-hour dive, I'm not sure I'd pay several hundred dollars for that, but maybe. The threshold of usefulness seems about there to me, at least at first approach.

As for the Honeywell Dead Reckoning Device - it seems to be essentially a very smart pedometer using GPS for step calibration, I think. There's a brief mention in the video presentation from the study. They brainstormed the idea of sensing fin strokes to do something similar, but I don't think that translates well.

But thanks for the pointers to these. At least get people thinking about the problem, with the benefit of other's thoughts to build on. Maybe somebody can make this work.

Hi reefduffer, thanks for your kind comments.
I felt also wanting more after reading the report. The conclusions were, for me, very very light.
The pedometer approach seems not to be useful, as there is not such as a step in a diver, specially if there is some current where the diver is drifting.

I'm very interested in a device like this, as I, as many people mentioned here in SB, find hard to surface near the dive boat when the dive ends. Tracking the position, and course back to the dive boat, makes me deduct enjoyment to the dive.

The Honeywell device seems to be a good start point with better sensors than those found in smartphones.

 

[Lorenzoid]

Thanks, guys, for letting us know there is nothing new in this area. I am confident that if I keep checking this thread every year or two, eventually I will find that someone has developed a feasible inertial navigation system for scuba.

 

[Kharon]

I'm in a bit of a hurry right now but will PM you my answers to your survey shortly. While I would love to have even a rudamentary UWPS (underwater positioning system) as I find navigation in very low vis a problem I would bet the cost would be prohibitive for me.

 

[TwistedGray]

Desert Star Systems, the Company I own, already makes a system for finding your way back to the boat, DiveTracker Sport or Scout. The thought of towing a float with a GPS above you while diving just sounds...well 2 steps in the wrong direction...

 

[emoreira]

Desert Star Systems, the Company I own, already makes a system for finding your way back to the boat, DiveTracker Sport or Scout. The thought of towing a float with a GPS above you while diving just sounds...well 2 steps in the wrong direction...

Hi TwistedGray, your device consist of a pinger and a receiver. The pinger should be located in the boat, near the boat in a float or in the anchor line. Some captains do not allow those types of devices. Installing such a device without the captain's permission could be cumbersome. I would not feel comfortable if I have to hide to install the pinger.
The use of a standalone inertial system would be preferable.
GPS in a towed buoy ? No thanks.

 

[Lorenzoid]

Sounds like Desert Star's target customer would be dive boat operators, not divers. And the dive boat operator would lend each diver or buddy pair a receiver unit. Am I right?

Nope. Like Eduardo, I'm still waiting for an underwater personal inertial navigation device to be commercialized.

Hey engineers who may be working on developing such a device, if accuracy is the biggest impediment, I would be glad to have a device that at least can point me in the general direction of the boat within, say, a quadrant. That would be a start.

 

[danvolker]

Brownies is about to have a gps nav device that can be towed on a torpedo float--or one of their electric 3rd Lungs....

The problem with inertial units, without gps, is no way to effectively account for ocean currents..which could mean hundreds of yards, to miles in just an hour long dive, depending on where you are diving.


And then there is always the Radar reflective Flag, that any diver could hold up with a speargun or on a good safety sausage--all the boat would need is simple radar and a Garmin chart plotter..