How to Build an ASV/USV: Hull Design
This is the first in a series of posts where I will walk through the basics of designing and building an Autonomous Surface Vehicle (ASV), also known as an Unmanned Surface Vessel (USV) or more simply, an unmanned boat or robotic boat. In the world of autonomous vehicles, ASVs lag far behind aerial drones and tethered submersibles (ROVs) in terms of popularity, although we are starting to see a surge in ASV projects, both commercial and amateur. Hopefully this series of posts will benefit those of you trying to build your own ASVs, especially if you are an amateur. If you have a question about an ASV design that you're working on, send me an email! Your feedback will make these posts more useful for everybody.
Flooded Hull or Dry Hull?
The first issue to consider when designing your ASV is whether the hull should be flooded or dry. Let's start by describing these two options:
Flooded: The hull is allowed to flood with water. This may seem like a terrible idea to those of us accustomed to boats that people ride on, but for a robotic boat, it's not a bad idea. In a flooded hull, part of the internal volume of the hull consists of rigid foam, either cut out from blocks or molded into the hull. This foam makes the ASV unsinkable. The rest of the internal volume consists of the electronics, payloads, actuators, etc., which are housed in waterproof boxes and connected to each other with waterproof cables and connectors. Each one of these components has to be independently waterproof because they will be submerged in water most of the time. Note that whatever space isn't taken up by the electronics, payloads, etc. should be filled with foam. Why? Because any volume that gets flooded with water represents additional weight. In other words, it will make the ASV sit lower in the water and require more power to move. So with a flooded hull, you need to keep everything neatly packaged. Think Tetris.
Dry: This is the more traditional way to build a boat. There is a big volume inside the hull that is sealed off with some kind of waterproof hatch or lid. You can put all your electronics, payloads, etc. inside the dry area without having to worry about waterproofing either the components themselves or the cables and connectors used to connect them.
It seems like a no-brainer: use a dry hull. With a dry hull, you only have to waterproof the hull itself instead of waterproofing all the individual components (electronics, payloads, etc.). That will save you both time and money. But hold on a second... are you really going to be able to keep the inside of the hull dry? Is your "waterproof" hatch really going to be 100% waterproof? If you have a propeller shaft exiting the hull, how are you going to keep water from coming in where the propeller shaft penetrates the hull? What if you have a rudder linkage penetrating the hull? Or a cable running through the side of the hull and up to your GPS antenna on top of the boat? Or a cable going to a water quality sensor down underneath the hull? And what if you get a crack in your hull from hitting a log (or an iceberg)? The list of possible leak points goes on and on.
On a human-carrying boat, these problems are handled with a few different techniques:
Make the hull itself really thick so it won't get holes or cracks.
Put as many of the hull penetrations as possible above the waterline.
Install a bailing system of some kind to remove any water that does get into the hull.
Raise all the important components (like the engine and battery) above the floor of the hull so that they don't get wet even if some water does get in.
Don't operate the boat in a body of water where the waves can crash over the top of the boat and flood the hull.
Now go through the list above and ask yourself if these techniques will work for your particular ASV. If your ASV is relatively large and if it will be operating in calmer water, most of these techniques will probably work. Use a dry hull.
If your ASV is going out on the ocean and is less than 30 or 40 feet long, you have to assume that it will be flipped over multiple times, completely submerged in the water over and over, slammed by waves, twisted, warped, pounded, and smashed. In that case, it's going to be very hard to keep the inside of the hull dry. Use a flooded hull.
HULL SHAPE
This topic is a huge one, far too big to cover thoroughly here. But let's at least look at some basic principles:
Don't Reinvent the Wheel
Boats have been around for a really long time. Although we're still making incremental improvements to their shapes, for the most part humankind has already figured out the right hull shape for a given application. So unless you're an expert on hull design, you might want to copy what others have done instead of trying to reinvent the wheel. The key here is to understand which hull designs are good for which applications. For example, a canoe looks a lot different than a ski boat. But why? And if you're building an ASV, should it look more like the canoe or the ski boat? Hopefully the following paragraphs will help make the choice a little more clear.
Planing or Displacement?
Some hulls are designed for "planing." This is where the boat skims across the surface of the water instead of plowing through the water. Ski boats, for instance, have planing hulls: you can very clearly see them rise out of the water as they speed up and "get up on plane." Planing hulls have wide, relatively flat bottoms with a distinct squared-off stern (also known as the transom). Once the boat is up on plane, its weight is being supported by the force of the water hitting the bottom of the hull at high speed.
Planing is great because, at high speeds, it uses far less power than plowing through the water. But you have to have enough power to get up to speed in the first place. Ski boats have enough power with their 200-horsepower engines to get up on plane, but the 1/4-horsepower human engine in your typical canoe isn't going to cut it, which is why canoes don't plane.
Canoes are displacement hulls. This means that they support their weight by displacing water (google Archimedes' Principle). So even when you're paddling your canoe as fast as your muscles will allow, the canoe still sits low in the water and has to plow through the water. Because of this, the canoe has a pointy bow and a pointy stern to try to reduce the drag as much as possible. And the bottom of the canoe is typically rounded instead of flat, also in an attempt to make it move through the water as smoothly as possible.
So should your ASV have a planing hull or a displacement hull? Almost certainly a displacement hull. Very few are the ASVs that will have enough power to plane. Remember: displacement hulls move through the water instead of skimming over the surface of the water, so that hull needs to be SMOOTH (more on that later)!
Monohull or Multi-hull?
Multi-hulls, like catamarans and trimarans, are typically more stable than monohulls. That can be good or bad. It's good if you are carrying a payload that really wants to be level, and if you're on calm water. But it's bad if you flip over. Why? Because a multi-hull is not only stable when upright, but also when upside-down, which means that it won't want to flip back upright. If you're out on the open ocean, assume that your ASV will flip over. Now if it's a multi-hull, you may need some kind of elaborate system to get it back upright, like a pumped ballast system or even a float on a long, mechanical arm (sort of like how a turtle on its back uses its neck to flip upright... if you've never seen a turtle do this, you're missing out). Usually self-righting systems such as these are not worth the trouble.
Because multi-hulls typically have narrower hulls than monohulls, they will often be faster for a given amount of power. And in the case of sailboats, they are able to use their widely-spaced hulls to counteract the force on the sail instead of relying on a keel hanging down in the water. That's why catamaran sailboats are almost always faster than monohull sailboats, all else being equal. So if speed is your ultimate goal, a multi-hull may be the way to go. As a rule of thumb, though, an ocean-going ASV will be a monohull, while an ASV for lakes and harbors may be a multi-hull.
Keel or No Keel?
Even if you use a monohull, you may still need some type of keel to ensure that the ASV will be self-righting. But often you can make the keel do double-duty by putting the motor and maybe even batteries down in the keel. That also has the advantage of putting the propeller further below the surface, where the water is less turbulent. The keel may also be a good spot to put any underwater sensors you may have.
If you do use a keel, try to keep its surface area as small as possible (more on that later). The Scout transatlantic ASV (pictured here) is a good example of a monohull with a long, slender keel.
Sleek and Smooth or Boxy and Faceted?
Sleek and smooth is the way to go! The reason I bring up this fairly basic point is that I see a lot of hulls that are unnecessarily boxy and faceted: either the bow is too blunt or the stern is cut off or the sides and bottoms of the hull have all kinds of sharp angles on them. Take out your sanding block and make that hull smoother! Your ASV will go a lot faster and use a lot less energy. There are some very complicated equations behind the optimization of hull shapes, but you don't have to be a fluid dynamicist to know that smooth shapes move through the water better.
Frontal Area vs. Wetted Area
Frontal area is just what it sounds like: the area of the hull as viewed straight-on from the front (or from the rear, I guess). It's intuitive that a hull with a lot of frontal area, like a tugboat, will be difficult to move through the water.
What might be a little less obvious is the effect of wetted area on a hull's performance. Wetted area is the total surface area of your hull that is in contact with the water ("wet"). So a long, thin, rowing shell, for example, has a very small frontal area but a pretty significant amount of wetted area, due to its length.
The more wetted area, the slower your ASV will go, since all that wetted area generates friction with the water. But how significant is the friction from wetted area? Very! Let me tell a quick little story to illustrate:
Many years ago, I had to rescue my daughter's kite from a river (don't ask me how it got in the river... I'd rather not talk about that). I swam thirty or forty feet to where the kite was floating in the river, grabbed the tip of the kite, and started swimming back to shore, pulling the kite along the surface of the water. I assumed that I would be able to swim effortlessly while towing the kite, but was shocked to find that pulling the kite along the surface of the water was like trying to pull a sword out of a stone. I quickly became exhausted in the cold water and would have drowned if I hadn't abandoned the kite and used my little remaining strength to make it to shore.
What happened? Well, the kite was nearly a 2-dimensional shape, with almost no frontal area at all. But it had tons of wetted area. All that friction between the wetted area of the kite and the water made for a ton of drag.
So don't underestimate the importance of wetted area. Keep the wetted area as small as possible. If you put a keel on your ASV, design it so that the area of the keel (as viewed from the side) is as small as possible. Keep fins, rudders, and anything else that's sticking down into the water as small as practical. And in an effort to make your hull sleek and smooth, don't make it too long, as that will add wetted area.
From the standpoint of wetted area, the optimum cross-sectional shape for the bottom of a hull is a semi-circle. A hull that is semi-circular in cross-section will have the smallest amount of wetted area for a given displacement (assuming that the waterline is at the diameter of the semi-circle). If you look at a rowing shell, the cross-section is nearly a perfect semi-circle. Unfortunately, this shape gives almost no stability (which rowers know very well), so it may not be the "right" answer. But try to keep the wetted area as small as possible while meeting your other requirements for stability, payload carrying ability, etc.
Fineness Ratio, Wave Drag, and Hull Speed
Related to the frontal area and wetted area is the concept of fineness ratio. The fineness ratio is simply the length of the hull divided by the maximum width of the hull: a rowing shell has a high fineness ratio (or is very fine) while a tugboat has a low fineness ratio (or is not very fine). Generally, a hull with a high fineness ratio will have a smaller frontal area and a larger wetted area, while the opposite will be true for a hull with a low fineness ratio. But there's more significance to fineness ratio than just frontal area and wetted area. The fineness ratio also comes into play when we start considering "wave drag."
Wave drag is a difficult concept to understand, but basically it's the drag associated with the waves, or wake, that a boat creates as it moves through the water. At slow speeds, the wave drag is almost negligible, but it quickly increases in a non-linear fashion as speed increases. Thus, wave drag creates a kind of speed limit for boats. This speed limit is called the "hull speed." Expressed in knots, the hull speed is approximately 1.34 * sqrt(hull length in feet).
Hull speed is not an absolute speed limit, but rather the speed at which the wave drag starts to increase very rapidly. It is roughly equivalent to the sound barrier for airplanes. If you're trying to go faster than the hull speed, it's going to take a huge amount of power. By the way, this is why fast boats like to plane: when a hull is planing, it doesn't create waves in the same way, and therefore is not subject to the speed limit imposed by the hull speed (but as we said before, planing requires a lot of power, too, just less power than if it were trying to plow through the water).
Hull speed is purely a function of the length of the hull, so the longer your hull is, the faster you can go before the wave drag starts to increase to a ridiculous amount. Similarly, the shorter your hull, the lower the speed at which wave drag starts to increase ridiculously. This is bad news for small boats like most ASVs. Example: if your ASV is eight feet long, the hull speed is 1.34*sqrt(8) = 3.8 knots. This implies that you're out of luck if you want to go 5 knots or 10 knots.
However, you can get around this problem by increasing the fineness ratio of your hull. A very fine hull (long and narrow) does not experience wave drag to the same extent that a short, wide hull does. As a rule of thumb, a hull with a fineness ratio of 12 or higher will experience significantly less wave drag than a less-fine hull. So if you absolutely must go faster than hull speed, use a very fine hull.
HULL CONSTRUCTION
This is another huge topic, but let's just cover the options at a high level:
Composites
The term "composite" refers to either fiberglass (which is a composite of fiberglass cloth and epoxy or polyester resin) or carbon fiber (which is a composite of carbon fiber cloth and epoxy resin). In my opinion, this is the best construction method for most ASVs. The only exception to that would be if you are making your ASVs in large quantities, in which case you might use a molded plastic hull.
Composite hulls can easily be made in almost any shape. The materials used in composite construction have no problem being submerged in water for years and years. And most importantly, composite construction is light, stiff, and strong.
The most professional way to build a composite hull is to first build a female mold for the hull. To build a female mold, you first make a "master" in the shape of the hull out of whatever material is most practical (usually foam covered in fiberglass or MDF sprayed with several coats of hard resin). Then you sand the surface of the master until it's super smooth and apply wax to the surface. Then you put several layers of fiberglass over the master. Once the fiberglass has cured, you pop the fiberglass shell off the master and you are left with a female mold from which you can now mold dozens of hulls. Piece of cake, right? Well, not really. There are many steps involved, a lot of material consumed, a lot of smelly fumes, and a lot of elbow grease. But yes, it is both possible and extremely rewarding. There are also plenty of professionals out there that will happily build you a mold, although they are going to charge a good amount of money for it.
You can also build composite hulls without a mold. The simplest technique is to carve the shape of the hull out of blue insulation foam from Home Depot. But how do you get the right shape? Unless you have access to a CNC router, you will have to do it by hand. Make a series of templates out of thin MDF or thick paper and sandwich them in between blocks of foam. Then carve and sand the foam away until you hit the edges of the templates. The picture here shows the SeaCharger hull after I carved and sanded the foam down to the templates (you can see the edges of the templates as faint stripes along the length of the hull). Once the foam is carved and sanded to the right shape, you then apply layers of fiberglass or carbon fiber. After the fiberglass cures, you can carve out pockets in the foam for whatever needs to go inside the hull. You can even dissolve the foam away completely with chemicals, leaving just the composite shell.
When building your composite hull, remember that you're going to want a few "hard points" in the hull where you can attach things like rudders, keels, etc. A hard point could be a piece of aluminum, brass, stainless steel, or thick fiberglass bonded into the hull under the skin, which can then be drilled or tapped to accept a screw.
When choosing between fiberglass and carbon fiber, keep in mind that metals that are in contact with carbon fiber in the presence of seawater will corrode, often quite dramatically (see this TechTip for more info). So you will have to electrically isolate any screws or other metal parts from the carbon fiber. Also, carbon fiber is more expensive and more brittle than fiberglass. On the plus side, it is lighter and stiffer than fiberglass. If you were building an airplane, it would be a no-brainer: use carbon fiber. For an ASV, it's a bit of a toss-up.
Wood
Wood is nature's composite. It is beautiful, strong, and easy to work with. I'd absolutely love to see more ASVs made out of wood (actually, I'd love to see even one ASV made out of wood). But wood requires extreme care in both the selection of the particular type of wood and in the waterproofing technique applied to the wood. And when all is said and done, it'll never be as light as a composite hull.
Welded Aluminum
This certainly makes for a durable hull. However, most people do not have the skills or tools to make a welded aluminum hull. Also, it's difficult to make a welded aluminum hull as smooth and sleek as it should be (aluminum hulls tend to be faceted). If you know your hull is going to be battered by logs or scraped across the rocks, this may be a good option. Otherwise, a composite hull is probably the way to go.
Plastic Kayak Hulls
You may be able to get away with re-purposing a plastic kayak hull as your ASV hull. If the kayak hull meets your other requirements, go for it! Plastic kayak hulls are usually made from roto-molded plastic and are extremely durable. They may not meet some of your other needs, however.
Summary
There are millions of ways to design and build an ASV hull. The key is to understand your requirements (does it need to be self-righting, is energy efficiency an issue, what tools are available for building it, etc.) and to weigh the options accordingly. Hopefully this post helps you to at least know what questions to ask. Good luck!