General Dynamics Mission Systems (GD) has opened a new Knifefish UUV (unmanned underwater vehicle) manufacturing facility at its Taunton, Massachusetts site.(1)
As a reminder, Knifefish is intended to be part of the LCS mine countermeasures (MCM) module and will be used to search for buried or bottom mines. Suspect mine locations and data are recorded to the UUV’s onboard data storage module for later upload to the host LCS for analysis. More recently, a data transmission capability has been added to try to streamline the data transfer process.
 |
| Knifefish UUV |
Knifefish is a 21” diameter, torpedo shaped UUV with a length of around 19 ft and a weight of 1700 lb. It uses a low frequency, broadband, synthetic aperture sonar. Knifefish is capable of operating for around 16 hrs (3) which, depending on the distance to and from the area of operation allows for perhaps 10-12 hrs of actual search operations. Its operational speed is a maximum of 4.5 kts(4).
GD is currently building five systems, with each system consisting of two UUVs, launch and recovery equipment, and control equipment.(1) The Navy is planning to procure 48 Knifefish UUVs with two allocated to each of 24 MCM modules the Navy hopes to buy.
We’ve often discussed the need to produce weapons and systems quickly during a time of war. So, what’s the production rate of the Knifefish?
The facility, approximately 8,000 square feet in size, will be capable of producing one system per month after receiving long-lead-time materials, said Craig Regnier, the manufacturing operations manager at the Taunton plant. (2)
One system per month? I hope we have a very slow war!
Unfortunately, Knifefish epitomizes the problem with the LCS MCM module. Knifefish is not a one-step mine disposal system as you might expect. Instead, Knifefish is just the first step in a multi-step process that requires a great deal of time. The major problem is that detection is not even remotely a real-time operation. The Knifefish UUV runs its scan pattern (at 4.5 kts!) and is then recovered aboard the LCS where the data is then analyzed. Carlo Zaffanella, vice president and general manager of the GDMS Maritime and Strategic Systems business noted this problem and identified it as something he wants to improve.
The next step is finding ways to make UUVs operationally more useful. A top focus here, he said, is the ability to analyze data while the UUV is still in the water, for something closer to real-time threat identification.
“Clearly, you would like to get to where more of that analysis were possible in real time, or at least as close to real time as you can make it,” Zaffanella told reporters. “If we could get the devices to provide not just essentially a map of what’s out there but perhaps detects or even tracks and say, ‘This is really what you want to be concerned about in real time,’ then the operational utility will go up.” (1)
So … even the manufacturer recognizes that the UUV has only limited utility and yet the Navy still wants it. Hmm …
The overall LCS MCM process involves multiple initial scans by various methods, followed by confirmation scans, followed by individual mine disposal. The mine clearance rate is something on the order of 1 per hour or less which is far from being combat-useful. For leisurely, peacetime clearance operations that can span weeks or months, that process might be fine but for combat operations that clearance rate is a non-starter.
Setting that aside, do you find the following as disturbing as I did?
Capt. Gus Weekes, the Navy’s LCS mission modules program manager, said he had watched GD staff launch a Knifefish at the waterfront in Quincy earlier in the week. During the testing, he said the UUV experienced a failure but praised the company for its ability to recover the drone despite the issue. (2)
So, let me get this straight. At a planned PR showcase event, where the UUV undoubtedly was exquisitely fine-tuned for the event, the UUV fails and the Navy program manager praises it? Shouldn’t this raise some red flags? Shouldn’t this prompt suspicion and criticism instead of praise?
Unfortunately, the current trend is that the Navy routinely accepts delivery of damaged and incomplete ships and equipment. Now, we’re extending that trend to praising malfunctioning MCM components. Shouldn’t the Navy have slammed the brakes on the program instead of praising it? This is how you get EMALS and AGS and LCS failures – by not questioning and holding manufacturers accountable. I’d have suspended all contracts and payments and demanded proof of success before resuming production.
This is yet another system whose effectiveness and value seem exceedingly questionable. Time after time, the Navy rides questionable systems right down the drain until they’re total failures instead of cutting their losses and terminating questionable systems early on. This is how we wound up with the LCS, Zumwalt, and so many other useless systems.
Even if Knifefish worked perfectly, it’s not suitable or effective for combat mine clearance operations. It’s time to cut the cord on this one.
_______________________________________
(1)Defense News website, “General Dynamics opens new unmanned underwater vehicle manufacturing center”, Megan Eckstein, 16-Aug-2021,
(2)Breaking Defense website, “GD Mission Systems Launches Knifefish Production Facility ”, Justin Katz, 17-Aug-2021,
https://breakingdefense.com/2021/08/gd-mission-systems-launches-knifefish-production-facility/
(3)https://www.naval-technology.com/projects/knifefish-unmanned-undersea-vehicle-uuv/
(4)https://auvac.org/files/uploads/platform_pdf/bluefin-21-product-sheet.pdf
The USN pretty much ignores mine warfare, except those occasions when it rears its ugly head, for which we are typically unprepared--Wonsan, End Sweep. Mine warfare is pretty much considered a career-limiting skill set, so very few senior officers have any useful knowledge of the subject. That is how we end up spending so much time and money on stupid ideas. That is also how we end up in the situation described by RADM Allan Smith at Wonsan, "We have lost control of the seas to a nation without a navy, using pre-World War I weapons, laid by vessels that were utilized at the time of the birth of Christ."
ReplyDeleteOur potential enemies--Russia, China, and Iran--do not ignore it. Neither do the North Sea countries--Germany, Holland, Belgium, France, UK--and Italy.
The saddest thing is that mines may be the most cost-effective weapons around. But they aren't shiny and sexy, so we ignore them until it is too late.
The only thing that rivals our shortcomings in MCM is our near total lack of effective offensive mine warfare!
DeleteOur mine inventory is somewhere between small and non-existent. Our mines are simplistic in the extreme - aerial bombs as mines??? Our delivery (mine laying) capability is non-existent and what there is, is almost never exercised.
The Royal Navy hasn't had naval mines in it's inventory for 30 years.
DeleteWhile for MCM is in the same class as the US with hardly any MCM ships and uses slow single use ROV's to destroy mines one at a time with only 4 ROV's per ship that gives the RN the ability to destroy 44 mines if all RN MCM's are in the same area then the mines win.
The really sad thing is that for about the amount of money we wasted on one Zumwalt, we could have a pretty impressive mining and mine countermeasures (MCM) capability.
DeleteThere are all sorts of influence triggers for mines—magnetic, acoustic, pressure, combination. You can almost pick the hull number of the ship you want to sink. It’s known technology, we just need to commit to building them.
On the MCM side there are basically two approaches—sweeping and hunting. Sweeping covers an area much faster, but gives you a confidence level rather than certainty. Hunting can ultimately achieve near certainty, but takes forever to get there.
I am intrigued by ComNavOps’s proposal that I have nicknamed “Wild Walrus,” where you would just send a bunch of unmanned vehicles down a channel programmed to blow up anything they find. This is kind of an analogue to hunting without wasting the classification time. Awesome thing, shoot something. You’d probably blow up a lot of buoy anchors and rocks and other non-mines, but if you sent enough of them through to get all the mines then you wouldn’t care.
The idea behind helo sweeps (and the German drone sweeps) is that we don’t want to risk live humans on sinkable ships in minefields.
I’ve thought about two kinds of MCM vessels. One would be like a small LSD/LPD and would carry 2-3 helos on the flight deck and 3-5 helo sweep sleds and 4-6 drone sweepers in the well deck. The ship would be able to deploy, control, and recover the helps and sleds, the drones, and their sweep gear. The second ship type would be a mine hunter, maybe like what Holland and Belgium are building (although I’ve seen some reports that suggest they are going with steel hulls and I would much prefer GRP or something else non-magnetic on anything that is actually going into a minefield). Both ships would also have the capability to launch large numbers of Wild Walrus.
Upon encountering a minefield, do the Wild Walrus saturation attack first, then the helo/drone sweeps, then hunt to clear the area completely. Whatever amphibious or other operation was planned for the area would commence when the commander’s desired confidence level was reached.
“Awesome thing, shoot something” in the above should be “See something, destroy something.” The idea as I understand it is blanket coverage with enough UUVs that you don’t care about discriminating between mines and junk.
Delete"hunting without wasting the classification time."
DeleteCombat mine clearing is all about speed. Classification is a luxury that cannot be indulged in combat. Besides, who really cares what it was? Blow it up and keep going! The desire for classification is a product of the technology fixation (do it because we can, not because it's useful).
In case anyone was interested in the quote from RADM Smith about the WONSAN Korea action..... https://arsof-history.org/articles/v9n1_cia_paramilitary_sb_nemesis.html
DeleteI'm wondering WTF happened to the mine-clearing sleds the MH-53E Sea Dragons used to pull. Are airborne mine-clearing techniques now as pointless as dive-bombing in the 21st century, maybe due to enemy air defenses? If not, shouldn't the USN modify some of its new CH-53K King Stallions to pull the sleds, or design a lightweight sled its SH-60 and MH-60 Sea Hawks can pull?
ReplyDeleteI believe there still may be some although, like the MH-53E's that pull them, they're getting long in the tooth.
DeleteThe Navy is developing the "Unmanned Influence Sweep System" for the LCS mine warfare package, which uses an unmanned surface vehicle to pull something that seems like at least part of this. Not sure if it's the full function.
I do recall there was a scandal early in the LCS Mine package development where there was something to be towed by the SH60 but turned out to be too heavy for the SH60. Perhaps it was this sort of thing. And of course the CH-53K is probably too heavy for the LCS.
"I'm wondering WTF happened to the mine-clearing sleds the MH-53E Sea Dragons used"
DeleteThe problem is that I'm not aware that sleds or sweeps are effective, at all, against modern smart mines. To the best of my knowledge, there have been no realistic tests. As best I can ascertain, sweeps will be of limited effectiveness against modern mines. We have criminally ignored mine countermeasures and what effort we've put into it has been woefully misdirected.
"The problem is that I'm not aware that sleds or sweeps are effective, at all, against modern smart mines."
DeleteThey have always been limited in effectiveness because they have limited power, whatever can be carried on the sled. We used to joke that they had a "swept path that was 6 feet wide."
It's actually better than that, and if we put money into R&D we could probably do far better than that. But we'd rather spend it on Zumwalts.
The German minesweeping drones carry larger sweep generators and as a result are more effective. The helos can cover more ground faster. I'd like to see the helo sweep with some kind of side scan sonar hanging off the sled.
What was wrong with the AN/SLQ-48 vehicle used by the old Avenger ships, which I believe does both identification AND elimination? Why does it need to be replaced by something that does only half the job?
ReplyDeleteThe 'problem' with the -48 is that it is a one-at-a-time system which is incredibly slow, though semi-effective (it reportedly has trouble with some types of mines). We've lost sight of the idea of combat MCM which is all about speed. The only way to achieve speed is via sweeping and that is a questionable proposition with modern smart mines.
DeleteWe desperately need to conduct some fundamental testing and research to see if sweeping is even viable today. The LCS program is trying to develop the Unmanned Influence Sweep System (UISS) but with no evidence that it will actually do the job, as far as I'm aware.
This thing is painfully slow at it's job! I don't have a big issue with USN trying out and working on the technology, my issue is why USN has already committed to putting into serial production (1 a month is serial production?!?ok....sarc) and 48 unit buy when its obvious there's still plenty of issues!
ReplyDeleteSomething IMHO that has gone wayyyy over board inside DoD in last couple of years is this mindset inside military and civilian contractors I think have started to really abuse is "we need to get XYZ weapon system ASAP to the warrior so they can play with it"....since when is putting half ass baked ideas and level 2 or 3 not working most of the time technology a good idea to give to some 18 to 20 years old? Why is the warrior supposed to figure this out?!? Especially when it's obvious that the weapon system in this particular case has many flaws and is unreliable?!? What are you really gaining here apart from the fact that it sure looks like the manufacturer is just going to use this to pretend to Congress that it kind of works and is in service?!? DoD really shouldn't allow this, only in some extreme cases!, if not, IMO a weapon system should only reach the warrior when it pretty much works and achieved a decent level of reliability AND they should be given a good grasp of how to operate the system! Warriors should just provide like the last 10% of new knowledge and employment of system. Not "let them play with it and come back and tell us how to use it..."
Agree 100%!! We've gotten completely off the traditional path from warfare capability gap to operational system/weapon/vessel. A solid CONOP is the nail consistently missed by the military/industry team hammer. I believe that we could still build valid, effective systems, but only if we direct the process properly, which is somthing we arent doing. We aren't identifying the capability gaps (actually many are quite obvious, we just arent moving logically to rectify them). We aren't looking at the simplest solutions, but instead, looking to use tech when its unnecessary/ overly expensive for the return. MCM is a great example of this. We know how to deal with mines, although the institutional knowledge is mostly forgotten. Instead of building large numbers of effective-if-not-perfect platforms, we're trying to go with exquisite, unmanned, zero-human casualty systems that will be painfully slow and therefore ineffective in actual combat. We've fallen into the trap of building a King Tiger when what we need are a dozen Shermans.
Delete"This thing is painfully slow at it's job!"
DeleteThe mystery is why the Navy would commit to something that, even if it worked perfectly, is too slow to be effective or useful in combat? This should have been aborted at the back of the napkin stage.
Unmanned mine sweepers have been around for a while. Following link is a historical event of unmanned mine sweeper during the Vietnam War.
ReplyDeletehttps://www.globalsecurity.org/military/world/china/ship-mine-vn.htm
Don't know when LCS will have all mine sweeper modules available, not just this one.
"Don't know when LCS will have all mine sweeper modules available, not just this one."
DeleteNever.
Doesn't China alone possess 70,000-100,000 mines depending on who you read? Iran and Russia are the same.
ReplyDeleteImagine if china put half dozen mines on each of there "PLAN" fishing boats and told them to salt the seas... No one could get close to Taiwan or any of the contested islands.
Delete"No one could get close to Taiwan or any of the contested islands."
DeleteI wonder if Commandant Berger has considered that as he pushes his forward base / forward missile concept?
I wonder what he has considered.
DeleteI'm not a proponent of unmanned, except for mine clearing.
ReplyDeleteI'd like to have something that can be unleashed in an area and have it cruise around destroying mines like an oceanic Roomba.
Lutefisk
An oceanic Roomba sounds good... Have them emit the acoustic signatures of various warships and be mine finding kamikazes!!
DeleteOf course theyd be expendable and built by the (tens of?) thousands...
"cruise around destroying mines"
DeleteWhat method do you envision it would use to destroy them?
"What method do you envision it would use to destroy them?"
ReplyDeleteMy understanding of mines and mine-sweeping is pretty basic, so please go easy if I'm saying something ridiculous.
I would think that the best method would be an unmanned vehicle (I like the idea of a torpedo-ish UUV) that mimics a ship.
I would assume that it would be possible to have such a vehicle emit an acoustic or magnetic signature similar to a ship's.
They would then follow a pattern, maybe a pre-designed pattern, or possibly random in nature.
This should clear the minefield.
But it would require the drones to be inexpensive, as a large number would be lost in the process.
They would need to be under the surface so that they aren't easily destroyed by the mine-laying navy/air force.
It would also be a slow process. Not something that could be done ahead of a task force, for example.
As I think through this, however, maybe dozens of these could be lined up to plow through a lane ahead of a task force/convoy, clearing a path?
Otherwise, I would envision this as something done to clear an area that is not an active zone, but may be in the future.
Lutefisk
That was my thought. While I dont know the exact criteria for triggering enemy mines, Id assume that sound is one of them. My dad had brought home a couple "practice torpedoes" from LBNS in the 70s. They were used for training shipboard ASW teams. While I was a kid and the details are foggy, I recall them "making the noise of a Soviet submarine", and being disposable. They were aluminum, with charging and test ports. I recall them having a soluable "plug" that dissolved when dropped into water, activating it. They were about 4ft long, 5in in diameter, and about 30lbs. They looked great polished and hanging over the mantle, but I digress... For sound activated mines this would be a great start, although being battery powered I wonder about their range, and assume we'd need it to be longer,but Id also assume battery tech has evolved enough to get the desired range without making it significantly larger. They were small and light enough that a good amount of them could be carried and air dropped.
DeleteNow, Im not sure how we'd trigger influence mines without that metallic mass, possibly a field generator(?) and that may require a bigger unit for a larger power source. Honestly, the older, dumb mines may in fact be a bigger challenge, requiring manual "sweeping".
Im sure many modern mines are complex and potentially discriminating, so creating little suicidal decoys won't be super easy, but Id think that creating economical units in large quantities, that can mimic one or more values for effective MCM isnt an impossible feat.
"This should clear the minefield."
DeleteWhat you've just described already exists and has for many decades. It's just basic influence sweeping where the sweeper tries to mimic a target ship. The only difference is that you've described an underwater version. Here's the challenges with doing that:
1. Power - A sweeper generates an acoustic or magnetic (or whatever) signature to trigger a mine. However, you don't want to be too close to the mine or your sweeper will be destroyed. So, the signal strength needs to be strong enough to be detected a good distance away from the sweeper. On a tow boat, you can install a fairly large generator. On a small (21", presumably) UUV, you can't install a very power source. Also, with a small power source you'll only sweep a small path. Sweeping just a few feet to either side doesn't help because it would take forever to clear a useful area.
2. Speed - Unless you're sweeping with a nuclear powered submarine, the speed is going to be very slow (3-7 kts is what is typically seen for UUVs) in order to have any useful endurance. Yes, a torpedo-ish UUV could travel at 30-60 kts but the endurance would be very short and you'd have to stop, recover, and refuel/recharge the UUV every ten minutes - a non-starter.
3. Smarts - A modern mine looks at more than one aspect of a target's signature to decide the validity of a target. This is a general problem with sweeping today. Can the sweep meet all the signature requirements to fool a modern mine? A small UUV with low power is going to be challenged to meet the signature requirements to trigger a mine. The small mass, for instance, combined with low power is going to have a hard time meeting the magnetic strength trigger point.
"Id think that creating economical units in large quantities, that can mimic one or more values for effective MCM isnt an impossible feat."
DeleteIt's a challenging task. Whether it's possible or not would become clear upon realistic testing … … … which for unfathomable reasons the Navy seems to have no interest in doing. I'm unaware of any realistic MCM testing. The lack of unrealistic testing, in general, is a recurring theme that I keep pointing out on this blog.
"What you've just described already exists and has for many decades. It's just basic influence sweeping where the sweeper tries to mimic a target ship."
ReplyDeleteI wonder if you could use magnetic scatterable sub-munitions to detonate mines.
In the 80's there was talk of doing that idea for anti-tank where the sub-munitions would float down and land on the tops of the tanks and blast through the thin armor.
Maybe something like that could be scattered over a minefield and the sub-munitions would flow to the mines and detonate on them.
Just a thought...
Lutefisk
(Don McCollor)[joining the discussion very late]...A mental exercise was thinking of a "smart mine" (much easier than their clearing). Single sensors do not indicate a ship -ignore them. Then do nothing until a random number of ships pass. The opponent is never certain the mines are cleared. Stalls movement almost as much as real mines...
ReplyDelete"Then do nothing until a random number of ships pass."
DeleteThese already exist.
Moving from the Knifefish specifically to autonomous ships in general, consider this column:
ReplyDeletehttps://www.forbes.com/sites/craighooper/2021/08/31/sea-buoys-solid-templates-in-navys-hyped-race-for-maritime-autonomy/?sh=7622f0ed43df
Basic idea: autonomous vessels do have promise, but at the moment they are just science projects, and will need a lot of work before they are ready to make a major contribution to our fleet. To get a flavor for the problem, consider our experience with our first (and most mature) autonomous marine vessel: the humble buoy.
Buoys are simple. No propulsion, they're just anchored to the ocean floor and float. They typically have a few simple sensors (for weather info, for example) and a simple radio to send the data to a base. And if you've ever seen one, you know they are built to be pretty robust.
Still, "data show that buoys have high failure rates, suffer a substantial amount of sensor attrition, and suggest that strategically important buoys often suffer above-average loss rates." This suggests that for unmanned ships, redundancy, in both numbers and onboard systems, will be critical.
Buoys also require a large logistical support base, which often requires technical helpers to travel to remote and inhospitable areas for extended periods. For example, "the Coast Guard dedicates a vast portion of their fleet—upwards of fifty cutters—towards buoy tending." Do our admirals realize this?
What does he suggest? Well, for one thing, getting autonomous vessels out to sea quickly, with simple capabilities, to help us understand what the sea does to these vessels, rather than having large "blocks" of designs with lots of capabilities, followed later by testing. Plus, we have to accept that there will be attrition, even in peacetime, so we'll have to buy more of these than we expect. And he suggests modifying contracting terms to try to fight hype by incentivizing demonstrated performance. "In a frothy field like autonomous systems, overall contractor veracity needs to be treated as a key performance indicator, on par with Congressional influence, industrial base concerns and parochial interests."
One problem that's kind of been implied here, but perhaps not stated as directly as need be, is that we haven't done anything even remotely approaching a major mining or MCM exercise or evolution in almost five decades.
ReplyDeleteWe need to do one or two annual Fleet Problems or Springtrains, they need to involve a substantial portion of the fleet, and mine warfare and MCM need to be a major part. Until we are doing those regularly, we are just kidding ourselves.
The LCS MCM effort is putting a great deal of emphasis (hope?) on the UISS. However, I'm unaware of any realistic testing of the sweep system to see if it can actually trigger modern mines. Have you heard of any testing?
DeleteThe only tests involving the UISS that I know of have been mechanical tests (launch, movement, communications, recovery). No mine triggering tests.
The UISS should be able to set off mines. The technology is pretty standard and proved. The extend of its ability to trigger mines will depend upon the size of the power source. The more power, the wider will be the potential swept path.
DeleteThe helicopter sweeps were limited by the size of the generator that could be mounted on the sled. The UISS can presumably carry a larger generator and therefore have a larger swept path.
With sufficient proper intelligence, the magnetic and acoustic generators can be set up to emulate the signatures that are set to detonate the mines. Of course if the mines carry a ship counter, then the amount of time required to sweep goes up significantly. The hard mine to sweep is the pressure mine, but what has been found is that wave action fortunately will set most of them off.
We don't have any way to sweep or hunt that is both totally effective and quick. For any MCM operation, it's going to be a tradeoff between assets, time, and the importance of the pending operation. If the pending operation has to go off in a hurry, then the operation commander will typically accept a lower level of confidence regarding mine countermeasures. Obviously, the more assets, the shorter time required. The LCS is almost unbelievably ill-suited to a mine countermeasures operation. My proposed MCM force would include a small-LSD/LPD-like ship that would carry 2-3 MCM helos on the flight deck, and 4-6 helo sweep sleds and 4-6 UISS-type devices in the well deck, and would have the capability to launch, recover, and control their operations.
And of course this mini-LSD/LPD mine countermeasures mother ship would have the ability to launch literally hundreds (because that's what it would take) of ComNavOps search and destroy UUVs that I have called Wild Walrus. The second MCM ship would be a somewhat smaller MHC that would go into the field and search and destroy anything left after the sweeping had achieved the required level of confidence. It would also be able to launch hundreds of ComNavOps's unmanned vehicles. With enough of both those types of ships, you would probably be in good shape to counter enemy mine threats.
DeleteAs far as the mining side, we probably need to upgrade the technology of our mine inventory. Between aircraft and submarines, we have decent number of mining platforms, but we really do need to upgrade our mines.
"we have decent number of mining platforms"
DeleteNot really. Minefields are typically laid in quantities of thousands. An aircraft or sub can lay dozens, at best. That either means many, many, many repeated runs or it means we need some other mine laying platform, such as a ship.
Yes, there can be occasional circumstances where a handful of mines would be useful that's not normally how they're used.
Aircraft are almost useless because they have to get far too close to the enemy to drop the mines in any useful locations. Against peer air defenses, that's suicide.
The study of WWII mine laying is interesting and instructive and it suggests that we have very little EFFECTIVE mine laying capability. We need to carefully game out where/how we want to lay mines and then decide what kinds of platforms we actually need. The answer, I'm sure, is ships. Perhaps stealthy, fast ships with large mine cargo capacities?
We also need better mines than aerial bombs with a semi-smart trigger.
Interestingly, I've never even heard of any offensive mine research being conducted. Just one of many mundane but war-winning capabilities the Navy refuses to pursue.
Be careful about letting wishful thinking drive expectations. UUVs take an incredible amount of systems engineering just to get in the water. The trade offs between mission duration, max speed, sensor type, communication, launch and recovery, etc. is driven to extremes very quickly by just minor changes in one parameter. A long mission duration quickly drives you to a large battery system. A high resolution sonar limits your speed and drives the size of the vehicle, along with the cost. High precision location drives INS costs up.
ReplyDeleteIn summation with seemingly small values of requirements you make the vehicle large > 12 inches diameter, long > 15 ft, heavy > 900 lbs, and expensive > $1M. And speed will dive the cost even worse as SAS do not work well at > 3.5 kts and 5 kts eats battery power like crazy.