In the News

Engineering and the Ocean Environment: Challenge and Opportunity

by Alissa Mallinson

Vast and seemingly impenetrable, the ocean inspires endless fascination. It is the topic of countless tales and adventures, from Captain Ahab’s pursuit of the Great White Whale to the discovery of the watery grave of the unsinkable Titanic.

The mysteries of the oceans’ depths and what lies beneath offer exciting challenges for engineers, who strive to develop new means to explore and utilize its resources. But why does the ocean generate such fascination and yet remain so unexplored?

“The ocean is very large,” says the William I. Koch Professor of Marine Technology and Director of the Center for Ocean Engineering Professor Michael Triantafyllou. “You can see that when you go looking for a crashed plane and can’t find it, and don’t even know where to look. There are parts of the Pacific Ocean that have never even been crossed scientifically since Captain Cook.

“And some people don’t recognize the ocean as interesting,” he continues. “For example, back in the ‘60s when the Alvin submersible was first dispatched, they wanted to go down and look at the deep parts of the Atlantic. But there were a lot of negative reactions around it. People asked, “What are you going to find? Why look at the bottom?” Well, they went there and they found the Mid-Atlantic Ridge, and all of a sudden Wegener’s tectonic plate theory was confirmed and changed the view of the planet.”

Indeed, it is only in the past few decades that researchers have really been able to inspect, investigate, and utilize the ocean environment. Its extent, depth, and extreme temperatures and pressure all present significant challenges to exploration technology.

At MIT, ocean engineering has always been a major element of our curriculum – notably the naval construction and engineering program 2N, which has produced many of the Navy’s top-ranking technical naval officers, and the naval architecture program, which produced several America’s Cup winners. The Department of Naval Architecture was established in 1893, and in 1976, it began a fruitful partnership with the Woods Hole Oceanographic Institution, creating a joint MIT-WHOI program in oceanographic engineering. In 1989, Professor Chryssostomos Chryssostomidis established the MIT Sea Grant Autonomous Underwater Vehicle (AUV) Laboratory, producing some of the first functional AUVs to become commercially successful. Several areas of mechanical engineering – such as mechanics, controls, design, optics,and robotics – play a large part in modern ocean engineering, and they all interface as we navigate the idea of responsible exploitation and protection of the ocean.

Many ocean engineering faculty in MechE have been at the forefront of ocean discovery and achievement, such as the program on Arctic acoustics that led to such fundamental discoveries as the first proof of Arctic Ocean warming; the Heard Island experiment, during which Professor Emeritus Arthur Baggeroer was part of a team that became the first to find, identify, and calculate average ocean temperature measurements; the introduction of data-driven ship design by Professor Jerry Milgram, leading to an America’s Cup win; the first marine biomimetic robot co-developed by Professor Triantafyllou; and ocean acoustic waveguide remote sensing (OAWRS), co-developed by Professor Nicholas Makris, which enables the observation and tracking of massive fish populations in their natural habitat and migratory patterns.

The oceans are utilized for everything from transportation (approximately 90% of the world’s transportation takes place by sea) and defense, to oil production, fishing, and entertainment. At MIT, we are keenly aware of our duty to utilize them as a resource while also understanding the impacts of that utilization.

“The ocean is a global system that needs to be thought of as a whole piece, not just parts,” says Professor Alexandra Techet. “There is so much of the ocean that we were not able to get to until technology allowed it. It is great that we are able to utilize our oceans, but at the same time, if we don’t protect them, there will be no more ocean to utilize.”

“The ideas of utilization and protection of the oceans go hand in hand,” adds Professor Henrik Schmidt, Director of the Lab for Marine Sensing Systems (LAMSS). “Whenever you start using or trying to exploit the oceans’ resources, you have to make sure you know what the impact will be. So we need to put the infrastructure in place that allows us to monitor what’s happening and take action if needed. Since 95% of the ocean is still unexplored, there’s still a lot we don’t understand.”

Professor Nicholas Makris’ acoustic imaging breakthrough in 2006 enabled a new means to look into the oceans. His OAWRS technique allowed researchers to take images of areas about 100 kilometers in diameter every 75 seconds. Compared to previous techniques, it used low-frequency sound waves that can travel far distances, providing a new way to track marine life and its migrations, and set the foundation for the use of acoustics as a means for gathering ocean data.

Researchers at MIT have also played key roles in developing underwater vehicles for ocean exploration. At first, they were large, expensive, and could only follow very simple directions, but the ability they offered to start exploring the deeper, less hospitable parts of the ocean was the foundation for all the investigation, responsible exploitation, and protection that came after. They were, among other things, an efficient way of reaching extreme ocean depths to gather samples and information that could be sent back to the surface.

Today, researchers are working on ways to send multiple AUVs out for exploration as a fleet, to gather data collaboratively and send it back immediately. But communications underwater have traditionally been a great challenge because electromagnetic signals only travel well underwater at very low frequencies, light attenuates rapidly, and the amount of information that can be transferred over acoustic channels is low.

Professor Franz Hover is looking at ways to give AUV fleets more sophisticated directions to allow them to communicate effectively with each other, something he likens to storm chasing.

“Here in the terrestrial zone,” he says, “we’re watching the weather very carefully: its winds and clouds, and physical properties. All those things are going on underwater too, through moving water masses with different temperatures, chemistries, and critters.

“On land, we have ubiquitous connectivity of agents; you can have wireless connectivity across miles and get very high coverage and good rates of information transfer. We’d like to have that underwater as well to monitor the oceans and go where things are exciting. There are important science, policy, offshore industry, and defense questions that you’d be able to answer if you had these observing capabilities.”

Hover envisions a group of mobile underwater vehicles that can communicate with each other, but more importantly, can develop and act on a global model of the situation at large, individually and collectively, reporting back to a dynamic control system that receives the data in real time and distributes commands based on a full understanding of the situation. “Underwater we’re going to pay for every single bit of information that passes acoustically between these vehicles,” says Professor Hover. “Agents don’t really have the ability to share all their information with each other or update each other very frequently. So what if the vehicles could exchange less information yet still follow the event they’re studying?”

One of the kayaks Professor Hover is using in his robotic control systems research.

Where Professor Hover’s solution to underwater observing systems is based on sophisticated controls communicated acoustically, Professor Henrik Schmidt is developing onboard intelligence and autonomy of AUVs based on data they gather acoustically.

He’s developing the infrastructure to observe and study the oceans by commanding his AUVs to map the ocean and track acoustic events one specific directive at a time, then training them to make an intelligent decision in real time about what to do with the data they gather.

For example, in the case of a missing airplane, says Professor Schmidt, normally a robot would be sent down to map areas using a lawnmower path. But because acoustic communications can’t transfer large amounts of information, the robot has to come up to the surface to send back its data, then wait for an above-water operator to analyze it and respond with instructions on where to hone in. Schmidt’s robots, on the other hand, are able to analyze sound underwater and make their own intelligent decisions about what to do with it.

“The underwater robots being sent to the bottom of the ocean – down to 5,000 meters in depth in some cases – have to be able to complete the mission of finding something, identifying what it is, and locating where it is accurately enough to pick it up or follow it, and that requires significantly more onboard intelligence,” he says.

“That’s where the artificial intelligence becomes such a key technology. We are essentially trying to clone expert understanding of underwater sound and put that into the robots, so that if they’re using sound for mapping or location purposes, they know when they see something abnormal, and can say, ‘Let me go look at it’ without waiting for an external command.”

Students learn to program AUVs to adaptively explore the marine environment, a core mission of Professor Schmidt’s lab.

But in oceans so vast, how do researchers choose the best routes for their robots? To answer that question, Professor Pierre Lermusiaux conducts ocean modeling research, particularly the characterization and prediction of uncertainty in ocean dynamics, to help optimize the paths of AUVs. In turn, data from these AUVs can be assimilated into Professor Lermusiaux’s model to help constrain his calculations, providing a greater degree of confidence in predicting data for regions where AUVs haven’t visited yet.

With these technological advancements in imaging, communications, and modeling, we are developing the tools we need to better explore and understand the oceans. Alongside exploratory tools, there is also a need for engineering technology that improves our operations in the ocean environment, addressing key societal needs such as transportation, defense, oil extraction, fishing, and disaster response.

With motivations such as this in mind, Professor Alexandra Techet has looked to biomimicry to investigate ways to improve the performance of underwater and air-sea vehicles. Professor Techet develops 3D imaging technology to study the physics behind the propulsive performance of accomplished sea swimmers and jumpers, which are able to gracefully maneuver in and out of the water.

“Salmon swim upstream and jump out of the water, whales breach, and archer fish can jump from a dead stop,” says Professor Techet. “They’re looking at their prey above the surface, and they go from zero velocity to shooting out of the water just by flipping their tail back and forth. How?

“We want to understand their propulsive performance, jumping capability, and maneuverability, and apply that knowledge to a vehicle underwater. We’re not necessarily going to build a mechanical fish because it would likely be too heavy to get out of the water, but we can understand the hydrodynamics required to propel something from either a slow speed or zero velocity out of the water.”

First she needs the tools to observe the hydrodynamic behavior of locomotion in water.

“Fully temporally and spatially resolved volumetric flow measurements are the next frontier in fluid mechanics,” says Professor Techet. “So the question is, ‘How can you do that experimentally?’ In my lab, we have developed a 3D particle image velocimetry (PIV) measurement tool that allows us to study higher speed flows and more unsteady problems.”

Professor Triantafyllou’s research on underwater vehicles has also been inspired by the mechanics and biology of marine creatures, particularly fish and seals, but his focus is on their ability to map the environment around them, the flows and eddies, by sensing changes in water pressure.

“The difference between fish and submarines,” he says, “is that a submarine does not sense what’s happening around the propeller or the rudder. This is not a capability we would even think about if we hadn’t noticed it in marine life.”

Using micro-electro-mechanical systems (MEMS), Professor Triantafyllou is able to emulate the sensing capabilities he’s observed in fish and seals in lightweight and cost-efficient ways.

“We were working on underwater robotic vehicles that looked like fish, and one of the things we wanted to explore was a way for such vehicles to extract energy from surrounding flows. We discovered that trout actually do this – they hide behind rock formations, using minute motions and taking energy from the rocks. We wondered how the trout knew where these wakes are, and it turns out it’s because of what we call a ‘lateral line.’”

An array of 10 MEMS velocity sensors emulating the lateral line of fish.

This “lateral line” is comprised of hundreds of tiny sensors on the side of a fish through which they can sense that an object is near, like when a truck passes by you on a highway and you sense a blast of pressure.

Professor Triantafyllou has been similarly inspired by the whiskers of seals. After conducting research on how they work, he discovered that they don’t shake unless they are affected by a change in pressure, but when they do, the seal knows that a fish has swum by and starts to pursue it.

Such underwater sensing capabilities could allow ships to detect an eddy forming under its hull from the drag of a sharp turn, slowing down the ship, and counteract it with opposing forces, or sense a current in the path of an AUV, giving it a chance to change course and avoid expending energy to fight it. This technology could also be used to locate objects or other transient events in the oceans, such as an oil plume.

The foreseeable future of the ocean presents a rich frontier of ocean engineering challenges to improve our ability to investigate, understand, and operate in this relatively unexplored system. But the more we are able to utilize the resources and opportunities the oceans have to offer, the more we are responsible for protecting them.