Two grey hulls bow-on at a berth, a concrete pier head with bollards and fenders between them, container cranes and a container ship beyond and a hillside town behind; below the waterline, the piles crusted with growth and rings of CS-320 creepers holding station around the piles and beneath the two hulls, over a rocky seabed.
Water · CS-310, CS-320 and KMM-01

Two Hulls.
And a Boat With Nobody On It.


One hull covers the distance on a screw. The other has no screw at all, creeps on a pulsed elastomeric bell, and rides in the first one's bay as a neutral cartridge. Above them both, KMM-01, a configuration rather than a third program, works the surface with nobody aboard. This is the family's front door: what each machine buys, and why the pair below is the product.

The family

One Hull Covers the Distance.
The Other Covers the Last Kilometre.


Two hulls swim, and the argument for them is the pair rather than either one alone. Neither is a smaller version of the other, and a third machine works the surface above them.

CS-310 is the survey-and-carrier hull. It is a torpedo-form body of revolution, 2.60 m overall over a Ø345 mm fairing at a fineness ratio of 7.5. Inside it, a ring-framed polymer pressure hull is the only dry space; below it, a flooded bay at the centre of buoyancy takes one cartridge at a time; at the tail, a shrouded screw sized for range rather than speed. It goes out, works a line and comes back with the picture.

CS-320 has no screw. A cast elastomeric bell contracts on a slow stroke, expels a ring of water through a Ø40 mm nozzle and refills on the recovery, over and over, for days. It creeps at 0.25 m/s, closes at 16.34 kg and 15.95 L wet, and its whole identity is being difficult to detect. It cannot cover an approach, and it is never asked to.

What holds the two together is a bay. CS-320 rides inside CS-310's 20 kg / 20 L wet bay as a neutral cartridge and is released at a surveyed point, so the expensive navigation stays on the larger hull and the smaller one begins its leg already close to the work. A buyer who needs both is buying one relationship, not two systems. Above the pair, KMM-01 is the uncrewed clearance workboat, a configuration of the mine clearance work rather than a third program: it stays on the surface, carries a flooded corridor between its own hulls, and lifts inert objects into cassettes under a supervisor ashore. Each program carries its own specification: CS-310, CS-320 and KMM-01.

The CS-320 on a clear white ground: a wide ribbed translucent bell above a dark cylindrical core, a pale payload cartridge held in cradle arms below it, then the ballast module and the drop weight at the foot.
The roster

The Water Family.
The Programs, One by One.


Two programs in the Water family — each with its mechanism, its doctrine and its deployment on its own page. The whole roster, all 13 programs across the fleet, is on the programs index.

Water · 2
CS-310 product render Water

CS-310

Persistent survey & carrier AUV — one moulded hull that surveys, carries and watches.

CS-320 product render Water

CS-320

A soft pulsed-jet payload carrier — slower than the current on purpose.

The medium

The Doctrine the Fleet Chooses,
the Sea Enforces.


No radio and no GPS in any Canadian Shield design that swims. Everywhere else in the fleet that is a decision taken against the grain of the market. Under water it is the operating condition, and it binds every side equally.

Seawater conductivity kills a radio signal outright a few metres down, the way rock kills it underground, and a satellite fix does not reach a vehicle that is under the surface. The rule the rest of the fleet holds by choice therefore costs this family nothing to hold. There is no receiver aboard either hull to jam and no navigation signal for anyone to spoof, and an adversary's jammer spends its power on a vehicle that was never listening.

What replaces the link is a wire and a plan. Where a person stays in the loop, the loop runs down an optical fibre paid out of a spool, which is a physical thread with nothing in the water to interfere with it. Where nobody is in the loop, human authority lives in the mission plan and the abort criteria, written at the shore station before the dive and held by the vehicle at mission grain rather than at joystick grain.

No acoustic device in the family is a control path. CS-310 carries a low-rate acoustic modem as a sparse check-in, abort and retask link and never as a control channel. On CS-320 a micro acoustic modem rides as an option rather than a baseline, because every transmission spends the signature budget the smaller hull exists to conserve. Silent running with no transmissions at all is a planned segment on one and the default on the other. The four-domain version of the doctrine is on Doctrine.

Doctrine
No radio. No GPS. The sea enforces it.

Human safety comes first, on every line. The hull goes into the water and the shore crew stays on the pier, and the abort on both platforms is mechanical and needs no electronics to work. Carriers, not munitions: no warhead, no fuze and no energetic material, in any cartridge, in any variant, for any customer.

Under a container berth, looking along a row of concrete piles crusted with growth: CS-320s holding station among the columns at every depth, one close in over a rocky seabed, and the hull of a container ship alongside at the upper left.
Why polymer

Build It for Cost,
and You Have Built It for Silence.


Both hulls are polymer-first, moulded on dies cut in the group's own shop. That is a manufacturing decision before it is anything else, and it is also the whole of the signature posture.

Moulding a whole vehicle in polymer is what makes it cheap to build in numbers, and a vehicle with no steel in it leaves no magnetic anomaly for an influence sensor to read and only a weak sonar return. One choice at the tooling stage buys both, and the manufacturing story and the low-signature story turn out to be the same story told twice.

On CS-310 the hard engineering concentrates in a single part, a ring-framed glass-filled nylon cylinder with two domed ends. Everything the water touches first is a free-flooded moulded shell that feels no pressure differential at all, so it can be light, inexpensive and quietly shaped. On CS-320 there is no metal structure at all: the propulsor, the fairing and the snag-compliance layer are one cast moulding, and it is the most-plastic platform in the fleet.

The obvious objection is mass, and the fair comparison answers it. Size an unstiffened 6061-T6 aluminium hull to the same safety factor at the same depth over the full unsupported length it actually has, rather than over a ring pitch it does not, and it comes out at 22.3 kg against the polymer's 24.6 kg. The non-metallic posture costs about 2.3 kg, roughly 2% of displacement. That is the price of a hull with no steel in it.

Depth at sea is hulls in the water rather than one exquisite vehicle, and a moulded hull off an owned die shop is what makes depth affordable. Where the dies are cut is on Manufacturing.

CS-310

Four Rings Are What Make
the Depth Rating Exist at All.


The pressure hull is Ø240 mm across a 12 mm wall over a 1,500 mm cylinder, with 8 mm hemispherical closures and four ring frames of 40 × 25 mm section at 300 mm pitch. Every one of those numbers is load-bearing.

A polymer tube under external pressure does not fail in hoop stress. It buckles, and buckling answers to stiffness rather than to strength. A polymer under sustained external pressure loses stiffness with time, and a real tube collapses below what perfect geometry predicts, so both knock-downs come off before anything is called a pass. Applied to a plain wall, the options fail a safety factor of 2 at 100 m until 19 mm, and by then the wall has begun to eat the payload.

Four rings divide the shell into short bays, and the effect is not marginal. The same shell unstiffened reaches 0.60 MPa after the knock-downs, a safety factor of 0.60 at the depth the hull is meant to work at, so without the frames the rating does not exist. Framed, the governing mode is inter-bay buckling at 2.73 MPa, a nominal collapse depth of 271 m, and a safety factor of 2.71 at the 100 m design depth.

What it all weighs is the end of the argument a buyer actually pays for. Carrying the same 20 kg payload to 100 m in polymer lands at about 113 kg wet, which is what the requirement costs once every part has a mass and a displaced volume beside it. That is a dolly or a davit and a shore crew of two rather than an unaided lift. The ring sizing, the end closures, the diameter sweep and the ledger line by line are on CS-310.

The CS-310 on a clear white ground, seen from the port bow: the black moulded nose dome, the olive mid-body with its lifting rail and ring bolts, the stern planes and the shrouded screw in its ring at the tail.
CS-320

No Screw at All.
A Cast Bell Does the Swimming.


A free-flooded elastomeric bell over a rigid polymer chassis, with a small dry canister inside holding everything that has to stay dry. The bell is the propulsor, the fairing and the snag-compliance layer in one moulding.

One stroke describes the machine. Half a litre of water expelled in 0.30 s through a Ø40 mm nozzle is a jet velocity of 1.33 m/s and 2.27 N of momentum thrust while the stroke runs, and mean thrust is that figure multiplied by the duty cycle. At 1.5 Hz on a 45% duty the mean is 1.02 N, which carries the vehicle at 0.25 m/s and no faster. Half a metre a second would want 3.8 times the mean thrust a credible bell stroke of this size delivers, and the shortfall survives a generous drag assumption.

So 0.25 m/s is the whole speed the machine has, and the mission set is sized around a creeper with no sprint in it. Everything else follows from the absence of a screw: no shaft to seal, no gearbox to whine, no blade tip to cavitate and no wake. The dry canister is Ø160 mm across an 8 mm wall over 450 mm with 6 mm domes, sitting in the short-cylinder regime where the end restraint carries load, and the selected wall clears a safety factor of 3.41 at the 30 m design depth for 0.28 kg more than the wall below it.

The energy budget is dominated by the hotel load and not by drag, so the battery is spent staying awake rather than moving. Range scales as hotel load to the power −2/3, and doubling the silent-running draw from 1.0 W to 2.0 W costs 37% of the vehicle's range. A watt taken out of the sleep state is worth more on this machine than any refinement of hull or propulsor. The bell as a part, the stroke, the ledger and the cradle are on CS-320.

The cast elastomeric bell close in: ribbed and translucent, flaring wide from the bolted collar that joins it to the dark core beneath.
A quay seen from the waterline: a moored hull and its mooring lines above, and below the surface the piled columns running away under the deck with CS-320s working along them.
The carriage

The Small Hull Is a Cartridge
in the Large One.


CS-320 closes at 16.34 kg and 15.95 L wet, inside CS-310's 20 kg / 20 L bay with 18% margin on mass and 20% on volume. That is a fit, and it is also the family's whole argument.

Carriage here is a payload-cartridge configuration, not a docking system. One creeper, retained in a cradle, released on a command the bay interface already carries. It is one-way: the mothership transits and releases, and the creeper runs the last leg and surfaces at a planned point for its own recovery. Because every cartridge in the family is neutrally buoyant by rule, releasing 16 kg from a 113 kg vehicle is not a buoyancy transient at the exact moment the larger hull is trying to hold position. It is nothing at all.

The relationship exists for an engineering reason, and it comes out of the small hull's own arithmetic. At 0.25 m/s in the 0.75 m/s set the design assumes, the creeper's speed over the ground is −0.50 m/s. It cannot make ground against its own design current. Drift transit works for getting there and cannot work for arriving, so the terminal leg is tide-timed to slack water, run down-current, or held to a low-current pocket, and it has to start close to the work.

Delivery to a surveyed release point is therefore what makes the creeper's delivered-precision tier load-bearing rather than decorative. In slack water a 2 km terminal leg at 0.25 m/s is 2.2 h, which the endurance carries comfortably: the mission closes on energy, and the tide window is what gates it. The expensive navigation stays on the mothership, and each hull is still independently launchable: CS-310 from a pier, a truck bed or a vessel of opportunity, CS-320 by hand from a small boat or the bank.

What a leg costs

A Leg Is Bought in Watts.
Most of Them Are Hotel Load.


Two things decide how far a hull gets: how fast it moves through the water, and what the instruments draw while it does. At working speed, on both hulls, the second is the larger of the two.

On CS-310 the two halves of the budget are drag and hotel load, and at survey speed the second is the larger. With an inertial navigator and a Doppler velocity log live, the best-range speed is 0.91 m/s, delivering 139 km over 42.7 h. On preset dead-reckoning with neither instrument powered it is 0.51 m/s for 391 km over 212.8 h. A 1.5 m/s cruise is a passage speed for when the clock matters rather than a norm.

HullSpeedWhat it buysThe assumption it carries
CS-3100.91 m/s42.7 h · 139 kmThe range optimum, inertial navigator and Doppler log live, still water
CS-3100.51 m/s212.8 h · 391 kmPreset dead-reckoning, neither instrument powered, still water
CS-3101.50 m/s20.1 h · 109 kmPassage speed, navigator live, still water
CS-3200.25 m/s59 h · 53 kmThe creep, at a 3.40 W total draw, still water
CS-3200.15 m/s133 h · 71 kmBest range at a 1.0 W silent-running hotel load, still water

Power is set by speed through the water and range by speed over the ground. The gap between the two is the current, and it is named on every row.

What the current does to a leg is the other half of it. Allowed to re-optimize against the set, CS-310 holds 1.06 m/s into a 0.25 m/s current for 0.81 m/s over the ground and 104 km, 75% of its still-water range; into 0.50 m/s it holds 1.26 m/s for 0.76 m/s and 75 km, 54%. Speeding up is the right answer to an adverse current, because time in the water costs hotel energy whatever the vehicle is doing.

A representative emplacement puts a shape on all of it: 15 km out, place the node, 15 km back in still water, at 9.2 h and 22% of the battery. The screw that buys those legs is audible, and that trade is accepted for the survey role. Range and payload sit at the mothership's end of the family and silence sits at the creeper's, which is the reason there are two hulls and not one.

The stern of the CS-310 on a clear ground: the bolted tail joint, the cruciform stern planes around the tail cone, and the shrouded screw turning inside its ring behind it.
One interface

One Bay, One Cradle,
and Every Cartridge Neutral.


Both hulls carry, and neither strikes. The interface is defined once on each — mount, connector, retention, release and the buoyancy bookkeeping — so a new cartridge is a tooling job rather than a redesign.

THE WET BAY

CS-310, at the centre of buoyancy

A free-flooded ventral bay taking one cartridge at a time, nominal 20 kg / 20 L, behind a define-once interface. A release from the centre of buoyancy disturbs trim minimally.

THE CRADLE

CS-320, one module at a time

A mechanical mount, a wet-mate connector and a release under the core, carrying one cartridge of about 2 kg and 2 L. It is a scaled member of the bay's interface family rather than a specification of its own.

NEUTRAL BY RULE

Every cartridge, in seawater

The vehicle trims the same whichever cartridge is fitted, and releasing one is trim-neutral at the moment the hull is trying to hold position. The cost is that every cartridge is ballasted at build, on a fixture and a scale.

WHAT GOES IN THEM

Instruments, and the list ends there

Sensor nodes, tags and beacons, acoustic relays, environmental samplers, inspection packages for hulls, piers, intakes and outfalls, and the creeper itself. Cartridge internals are the customer's or a partner's scope.

The boundary binds in every variant and for every customer. No warhead, no fuze, no energetic material and no terminal-effect payload, in any cartridge. Any proposal to arm a platform in this family is a counsel-first corporate question and never a product decision.

Two items carry export treatment of their own: the inertial navigator and Doppler velocity log that make up the mothership's top guidance tier, and the emplace-and-retrieve cartridge. Both belong on the table in a first conversation.

The abort

The One Device That Has to Work
Is the Simplest Thing Aboard.


Both hulls end the same way when something goes wrong. A weight held on a normally-energized coil drops, and the vehicle rises. Cutting the power is what releases it.

Every other circuit on both platforms fails safe by being switched off. This one is wired the other way round, so anything that takes the power away is the release rather than a failure of it. It is the one subsystem that has to work when nothing else does, and it is the least clever thing on either machine.

The weight is cast iron rather than lead on both hulls, 3.00 kg on CS-310 and 0.80 kg on CS-320, because every abort leaves it on the bottom. A vehicle abandoned in a harbour is an environmental failure before it is an economic one, and so is the weight it leaves behind. Nominal recovery is a planned rendezvous at the surface and off-nominal recovery is the drop weight; either way the hull, the battery and the data come home.

Human safety comes first, on every line, and on the water this is where it shows. The machine goes in and the crew stays out: two people with a dolly or a davit for the mothership, one person and a pair of hands for the creeper. Nobody works over the side of a vessel to launch either one, and no fault on either hull asks anybody to go into the water after it.

The lower half of the CS-320 on a clear ground: the underside of the core canister, a pale payload cartridge held in its cradle arms, the ballast module below that and the drop weight at the foot.
Guidance

The Three Tiers,
Scaled to Each Hull.


The family runs three tiers, scaled to each hull. A vehicle holding a line in a moving medium is guided or preset, and there is no ballistic tier at sea.

TIER 1

Preset dead-reckoning

Legs planned before the dive and run on magnetic heading, a depth cell, time and a current prediction. Nothing is commanded in the water and nothing aboard is listening. On the creeper the current term dominates the error budget, so the drift plan is the mission plan.

TIER 2

Fibre-guided, a person in the loop

Guidance over an optical fibre paid out of a spool. It is kindest here of anywhere in the fleet: no launch shock and walking-pace speeds. It is the close-work tier on the mothership, and on the creeper spool volume competes with the cradle, so it is the inspection tier there too.

TIER 3

The capable tier, on each hull's terms

On the survey hull, an inertial navigator corrected by a Doppler velocity log with bottom lock, bought rather than built and export-gated in its own right. On the creeper, delivered precision: the mothership carries the navigation and hands over a surveyed start point.

Read as a purchasing question, the tiers describe what is physically aboard, not what the software does. A preset hull has nothing fitted that could receive. A fibre-guided hull has a spool and a thread. The top tier on the mothership is two instruments with a part number and an export file, and the top tier on the creeper is a decision to let the larger hull carry them. The framework the whole fleet uses is on Technology.

One consequence matters to a buyer used to a different market. Neither hull has a fallback to a radio link when the fibre runs out, and neither carries a receiver fitted for a degraded mode. What the vehicle is told, it is told before it leaves, or it is told down a wire.

The family, counted

Both Hulls, Row by Row.
The Same Questions, Different Answers.


The two swimming hulls side by side, with the figures each one's engineering carries. KMM-01 is a surface vessel on a different scale, a configuration rather than a program, and it is counted on its own page.

CS-310CS-320
What it isThe survey-and-carrier hull: route survey, seabed awareness and coastal watchkeeping on long legs, and the carrier bus for the tools that do the close work.The creeper: a soft pulsed-jet payload carrier for the last, quietest kilometre.
Pressure boundaryØ240 mm × 12 mm over a 1,500 mm cylinder, 8 mm hemispherical closures, four 40 × 25 mm ring frames at 300 mmØ160 mm × 8 mm over a 450 mm canister, 6 mm domes
Design depth100 m, at a safety factor of 2.71 after creep and ovality30 m, at a safety factor of 3.41 after creep and ovality
Overall form2.60 m over a Ø345 mm fairing, a fineness ratio of 7.5A cast elastomeric bell over the core canister, the cradle and the ballast stack
Wet displacement113.12 kg16.34 kg · 15.95 L
PropulsionOne low-RPM shrouded screw behind four cruciform stern planes, driven through a magnetic couplingA cast bell pulsing between 0.5 and 2 Hz through a Ø40 mm nozzle. No screw, no shaft, no wake
Best range0.91 m/s for 42.7 h and 139 km0.15 m/s for 133 h and 71 km
PayloadOne cartridge at a time, nominal 20 kg / 20 L, at the centre of buoyancyOne cartridge at a time, about 2 kg and 2 L, in the cradle
Crossings of the pressure boundaryFive electrical, down from ten on a naive layoutTwo electrical and one water port
AbortA 3.00 kg cast-iron drop weight on a preloaded springA 0.80 kg cast-iron drop weight on a preloaded spring

Both hulls are polymer-first and moulded on the group's own tooling, and neither holds an energetic material in any variant of either one.

The CS-310 cut out on a clear ground, nose to the left: the moulded nose dome, the mid-body with its lifting rail, the ventral bay rail below, the stern planes and the shrouded screw.
The CS-320 cut out on a clear ground, seen from the front: the ribbed translucent bell flared wide over the dark core, the cartridge in its cradle arms, the ballast module and the drop weight below.

Reach and payload capacity sit at the CS-310 end of the family, and silence and the last kilometre sit at the CS-320 end. Read the middle column for a hull that goes out and comes back, and the right-hand one for a hull that arrives without being noticed. The row that joins them is the payload row, because the second hull fits in the first one's bay. The workboat above them, KMM-01, is a 13–14 m surface vessel and shares none of these rows.

Under a pier: a field of piled columns running into the distance with CS-320s among them, and one close in the foreground.
Fit and scope

What Both Hulls Refuse.
In Every Variant, for Every Customer.


The fastest way to test whether these hulls answer your problem is to read what they will not do. Every line below holds on both swimming hulls, in every variant of either one.

Who works this water, what a first conversation needs and how an enquiry is screened are on who it is for.

The conversation

One Door for the Whole Family.
A Person Writes Back.


Bring the approach, the depth band, the current you have to time the work to, and the job at the far end of the leg.

Four pages under this one carry the family in full. The pressure hull, the ring sizing, the ledger, the five crossings of the pressure boundary and the wet bay are on CS-310. The bell, the stroke, the drift arithmetic and the cradle are on CS-320. The corridor between the hulls, the cassettes and the supervision arrangement are on KMM-01. The buyer's view, what a configuration is scoped from and how a first conversation opens are on Who It Is For. The doctrine these pages sit under is on Doctrine, the whole roster is on the programs index, and the enterprise's other half, the removable operator for machines a customer already runs, is on CDNS Autonomy.

Write with the water you have to know and the work at the far end of it, and say whether you are scoping one hull or the relationship. A customer's requirements shape the platform, and the bay is where that shows up first. Every enquiry is screened before any discussion; international transfer is subject to Canadian government permits taken per shipment, export posture is counsel-first, and every design in the family is patent pending. The reply comes in writing, from a person.

All Water family designs are patent pending; international transfer is export-controlled and subject to Canadian government permits taken per shipment, and nothing on this page is an offer.