Autonomous Ocean Robots: Swimming in a Sea of Data
Ocean robots have become a veritable industry unto themselves, enabling marine data collection on a scale never before possible. Read on to learn about the long-duration ocean robots enabling this paradigm shift in marine data collection.
Traditionally, manned ships and buoys have met most of the maritime industry’s data collection and monitoring needs. But buoys are expensive to deploy and are stuck in one spot. Ships are expensive (around $50,000/day) and require a full crew to keep them operating, not to mention their emissions. Autonomous ocean robots on the other hand are like moving buoys; are cheaper than ships; keep people on shore out of harm’s way; and are far more environmentally friendly. Most importantly, you can leave them out at sea for up to a year and they won’t complain or file a union grievance.
An autonomous robot is one that operates by itself, or with very limited supervision. Think self-driving cars in the ocean. There are no tethers or cables connecting it to its driver like with an Remotely Operated Vehicle; in fact, the pilot of the vehicle, if there is one, may be thousands of miles away. Most autonomous robots are autonomous in a loose interpretation of the word: they are constantly surveying their surroundings and have the freedom to act independently, but parental control is often required. For example, an autonomous robot may know that there is a cargo ship bearing down it, but it might not know how to get out of the way.
There are several different types of autonomous ocean robots (see the bottom of this article for an explorable table), but they can be generally grouped as either a surface or underwater vehicle. Autonomous Surface Vehicles (ASVs) (sometimes referred to as Unmanned Surface Vehicles (USVs)) fall into the surface category while Autonomous Underwater Vehicles (AUVs), Remotely Operated Vehicles (ROVs), and Gliders are all within the (you guessed it) underwater category. Below I’ve plotted a sampling of both surface and underwater autonomous vehicles. I’ll be focusing on the surface robots for the rest of this post.
Long-duration ocean robots are a unique type of autonomous vehicles that generally fall in the surface category. They are designed for longer periods at sea (weeks to a year) and serve primarily as observation, measurement, and intermediary communication platforms. They are best suited for missions where frequent or constant communication is critical, and where data needs to be collected over both time and space.
These robots perform a variety of tasks: from the mundane, like measuring the temperature of seawater, to the more exotic, such as tracking great white sharks or enemy submarines. There are literally hundreds of different sensors that can be packed into these vehicles.
Why are surface vehicles better for long deployments? Two reasons. First: surface vehicles have more opportunities available to them for generating their own power (solar, wind, waves) to charge their batteries. Underwater these power sources disappear. This means an underwater vehicle will need to carry a large enough power source for the duration of its mission.
Second: sending and receiving wireless signals through the water is difficult. When a robot is operating out in the middle of the ocean satellite communication is pretty much the only option to transport data to shore, which means the robot needs to have a direct line of sight with the sky. This is quite tricky when it’s 1,000 meters or more underwater. Sure, underwater vehicles could surface, but then they are off mission and not collecting data. For this reason in particular, ASVs have found a niche role in communication networks, serving as middlemen between aerial and subsurface vehicles.
An ASV is only useful when its sensors are collecting data, so the robots are designed to reduce power needs wherever possible. Propulsion is one of the biggest power hogs out there, so it shouldn’t come as a surprise to learn that most ASV designers have shunned traditional electric motors for more sustainable methods. The biggest players in the long-duration ASV market use waves, wind, or buoyancy to keep their robots moving.
The Wave Glider, manufactured by Liquid Robotics, uses wave motion to propel itself. This two-part system is comprised of a float and an attached glider beneath it. On the glider there are pivoting slats that are oriented transversely down its length. When the float and glider encounter a wave, the float lifts the glider upwards as it approaches the crest of the wave. Instead of just moving straight up, the slats on the glider pivot and act like wings causing the glider to move upwards at an angle. Likewise, when the float hits the wave trough the glider sinks back down at an angle. By virtue of this saw-tooth pattern through the water column the Wave Glider moves forward.
Depending on wave heights and periods, the Wave Glider can reach speeds in excess of 2 knots, which isn’t much, but if you’re not in a rush it’s more than enough. The AutoNaut uses a similar principle and achieves slightly greater speeds (likely due to a more streamlined hull). Clearly, this propulsion method is dependent upon waves, so places with little wave activity are generally poor environments for robots like the Wave Glider or Autonaut (but each comes equipped with back-up electric propulsion.
The Datamaran (built by Autonomous Marine Systems) and Saildrone (company of same name) both use wind as their main propulsion method. These robots use a wingsail (think airplane wing), which is simply a rigid airfoil. Any sailor will be quick to note, sails must be trimmed to generate lift and propel the boat forward. How then is this accomplished on an unmanned robot?
To keep the sail at the correct angle relative to the wind both vehicles rely on a ‘tail’ that is attached a certain distance behind the trailing edge of the main sail. The tail looks like a miniaturized version of the mainsail, however it is adjusted to an ever-so-slightly different angle relative to the wind. The wind flowing over this tail causes a lift force, which applies a torque to the mainsail and causes it to rotate away from the direction of the wind. Now the mainsail generates a lift force of its own to propel the vehicle forward. The beauty of this method is that the tail always keeps the sail properly trimmed. Neat huh? These vehicles move quick: Saildrone claims its vehicle tops out at 18 knots, though average speed is closer to three.
The other propulsion technology that is well suited to long-duration ocean vehicles relies on buoyancy. Buoyancy Gliders, as they are called, are one of the few long-duration underwater vehicles. They rely on a small bladder to control their buoyancy, and thus height in the water column. Affixed to the outside of the vehicle are a set of wings. When the bladder is filled with seawater the vehicle becomes negatively buoyant and sinks. Instead of sinking straight down, the wings cause the vehicle to glide down at a very shallow angle. When the glider has reached the bottom of its downward path, the bladder is emptied, the vehicle becomes positively buoyant and then begins its ascent. Vehicles using this method of propulsion aren’t very fast, typically topping out at less than a knot.
Although these robots use their environment to generate power, they are also at the mercy of it. Even the longest lasting ASV is limited to at most a year. This is due to a couple reasons. First, nearly all of the long-duration surface vehicles rely on solar PV to keep their batteries charged. Sometimes there just aren’t enough sunny hours in the day to keep the robots and their sensors operating. String together a couple days of no sun, and now the robot barely has enough power to communicate with shore, let alone turn-on sensors.
Many of these robots have speeds less than a couple knots. So areas where ocean currents exceed 3 knots, like the Gulf Stream or the Loop Current in the Gulf of Mexico are not ideal places for an ASV. This is especially true when the robot needs to maintain its position over a certain area, or watch circle. If one of these vehicles hits a strong current it’s likely going for a long, long ride.
One of the biggest limitations to vehicle duration is actually biological growth. These vehicles operate in the upper reaches of the photic zone where sea life is attracted to their hulls like a fat kid to cake. Barnacles and algae accumulate on a vehicle’s hull over time (faster in tropical regions, slower in colder regions) and this added drag drastically reduces the performance of the vehicle. It can become so bad that the robot can’t maneuver and eventually turns into high-tech driftwood.
We can only manage what we measure. To help us better manage our oceans we need to measure them, but close to 95% of the ocean still remains unexplored. Long duration ASVs may not be the fastest or carry thousands of pounds of equipment, but they are persistent, cheap, and sustainable. They are enabling oceanographic data collection at a scale never before possible, providing us with a better understanding of how the oceans are changing and how we might better protect them.
Ocean Drone Data Table
|Company||Vehicle||Classification||Primary Propulsion||Power Generation||Vessel Length (m)||Vessel Breadth (m)||Vessel Draft (m)||Vessel Air Draft (m)||Vessel Weight (kg)||Operating Depth (m)||Avg Speed (knt)||Payload Weight (kg)||Continuous Power Supply (W)||Peak Power Supply (W)||Battery Storage (kWh)||Duration (days)|
|Liquid Robotics||Wave Glider SV3||ASV||Wave Flaps||Solar PV||2.13||1.42||0.21||1||150||0||1.8||45||5-20||360||0.9 - 4.5||365|
|Saildrone||Saildrone||ASV||Rigid Sail||Solar PV||7||2 (est)||2||4.6||544||0||4||113.4||5-10||200||365|
|Autonomous Marine Systems||Datamaran||ASV||Rigid Sail||Solar PV||2.5||1.7||0.2||85||1||2.4||25||2||1000||0.6 - 4.2||180|
|Boeing||Echo Voyager||UUV||4 Thrusters||Diesel Generator Engine||15.5||3 (est)||3.7 (est)||4.5 (est)||50000||3300||< 8||20000||Unspecified||18000||Unspecified||180|
|ASV||C-Enduro||ASV||Electric Propeller||Solar PV, Wind Turbine, Diesel Generator||4.2||2.4||0.4||2.8||350||0||3||50||2500||Unspecified||90|
|Autonaut||Autonaut||USV||Wave Foil||Solar PV||7||0.9||1||3.5||4||200||Unspecified||Unspecified||Unspecified||60|
|Riptide Autonomous Systems||Micro-UUV||AUV||Electric Propeller||N/A||1.83||0.12||0.12||N/A||16.3||200||4||Unspecified||16|
|Bluefin Robotics||Bluefin-21||AUV||Ducted Thruster||N/A||4.93||0.53||0.53||N/A||750||4500||3||Fixed payload, unknown if customizable||Unspecified||Unspecified||13.5||1|
|Hydroid||REMUS 6000||AUV||Electric Propeller||N/A||3.96||0.71||0.71||NA||862||6000||3||27||Unspecified||Unspecified||12||0.92|
|Offshore Sensing AS||Sailbuoy||ASV||Rigid Sail||Solar PV||2||0.52||0.57||.75 (est)||60||0||1.5||10||Unspecified||Unspecified||0.28 - 0.56||60|