Sat. May 21st, 2022


MIT self-propelled undulating swimmers

Assistant Professor Wim van Rees and his team have developed simulations of self-propelled undulating swimmers to better understand how fish-like deformable fins can improve the propulsion of underwater devices, seen here from above. Credit: Image lent by MIT van Rees Lab

WITH marine and mechanical engineers use advances in scientific data processing to solve the ocean’s many challenges and seize its potential.

There are few environments as irreconcilable as the ocean. Its unpredictable weather patterns and limitations in terms of communication have left much of the ocean unexplored and shrouded in mystery.

“The ocean is a fascinating environment with a number of current challenges such as microplastics, algae blooms, coral bleaching and rising temperatures,” said Wim van Rees, ABS Career Development Professor at MIT. “At the same time, the sea holds innumerable possibilities – from aquaculture to energy harvesting and exploration of the many marine animals we have not yet discovered.”

Ocean engineers and mechanical engineers, like van Rees, use advances in scientific data processing to solve the ocean’s many challenges and seize its potential. These scientists are developing technologies to better understand our oceans and how both organisms and man-made vehicles can move in them, from the micro-scale to the macro-scale.

Self-propelled undulating swimmers

Assistant Professor Wim van Rees and his team have developed simulations of self-propelled undulating swimmers to better understand how fish-like deformable fins can improve propulsion in underwater devices, here seen as two fish side by side. Credit: Image lent by MIT van Rees Lab

Bio-inspired underwater appliances

An intricate dance takes place while fish dart through the water. Flexible fins flap in water currents, leaving a trail of vortices in their wake.

“Fishing has intricate internal muscles to adapt to the precise shape of their bodies and fins. This allows them to propel themselves in many different ways, far beyond what any man-made vehicle can do in terms of maneuverability, agility or adaptability. , ”Explains van Rees.

According to van Rees, thanks to advances in additive manufacturing, optimization techniques and machine learning, we are closer than ever to replicating flexible and transforming fish fins for use in underwater robotics. As such, there is a greater need to understand how these soft fins affect propulsion.

Van Rees and his team develop and use numerical simulation approaches to explore the design space for underwater units that have an increase in degrees of freedom, for example due to fish-like, deformable fins.

Prediction of Loop Current Eddies

Graduate students Abhinav Gupta and Professor Pierre Lermusiaux have developed a new machine learning framework to help compensate for the lack of resolution or accuracy in existing dynamic system models. Their framework can be used for a variety of applications, including improved predictions of Loop Current vortices around oil rigs in the Gulf of Mexico. Credit: Image lent by MIT MSEAS Lab

These simulations help the team better understand the interplay between fluid and structural mechanics in the fish’s soft, flexible fins as they move through a fluid flow. As a result, they are able to better understand how deformations of the fin shape can damage or improve swimming performance. “By developing accurate numerical techniques and scalable parallel implementations, we can use supercomputers to solve what exactly is happening at this interface between flow and structure,” adds van Rees.

By combining its simulation algorithms for flexible subsea structures with optimization and machine learning techniques, van Rees aims to develop an automated design tool for a new generation of autonomous subsea units. This tool can help engineers and designers develop, for example, robotic fins and submarines that can smartly adapt their shape to better achieve their immediate operational goals – whether it’s swimming faster and more efficiently or performing maneuvering operations.

“We can use this optimization and AI to create inverted designs throughout the parameter space and create smart, adaptable devices from scratch, or use accurate individual simulations to identify the physical principles that determine why one form performs better than another, “explains van Rees.

Swarming algorithms for robotic vehicles

Like van Rees, lead researcher Michael Benjamin wants to improve the way vehicles maneuver through the water. In 2006, as a postdoc at MIT, Benjamin launched an open source software project for an autonomous rudder technology he developed. The software, which has been used by companies such as Sea Machines, BAE / Riptide, Thales UK and Rolls Royce, as well as the US Navy, uses a new method of multi-objective optimization. This optimization method, developed by Benjamin during his PhD work, allows a vehicle to choose the course, speed, depth and direction it needs to go in order to achieve several simultaneous goals.

Swarming algorithms for unmanned vehicles

Michael Benjamin has developed swarming algorithms that allow unmanned vehicles, such as those pictured, to disperse in an optimal distribution and avoid collisions. Credit: Michael Benjamin

Now Benjamin is taking this technology a step further by developing swarms and algorithms to avoid obstacles. These algorithms would allow dozens of unmanned vehicles to communicate with each other and explore a given part of the ocean.

To begin with, Benjamin looks at how best to disperse autonomous vehicles into the ocean.

“Let’s assume you want to launch 50 vehicles in a part of the Sea of ​​Japan. We want to know: Does it make sense to drop all 50 vehicles in one place, or have a mother ship drop them off at certain points in a given area? ”explains Benjamin.

He and his team have developed algorithms that answer this question. Using swarming technology, each vehicle periodically announces its location to other nearby vehicles. Benjamin’s software enables these vehicles to disperse in an optimal distribution for the part of the sea in which they operate.

Central to the success of the swarming vehicles is the ability to avoid collisions. Collision avoidance is complicated by international maritime regulations known as COLREGS or “Collision Regulations”. These rules determine which vehicles have the “weather” when crossing paths, posing a unique challenge to Benjamin’s swarming algorithms.

COLREGS is written from the perspective of avoiding another single contact, but Benjamin’s swarming algorithm should take into account several unpilated vehicles trying to avoid colliding with each other.

To tackle this problem, Benjamin and his team created a multi-object optimization algorithm that ranked specific maneuvers on a scale from zero to 100. A zero would be a direct collision, while 100 would mean that the vehicles completely avoid collision.

“Our software is the only marine software where multi-objective optimization is the basic mathematical basis for decision making,” says Benjamin.

While researchers like Benjamin and van Rees use machine learning and multi-objective optimization to address the complexity of vehicles moving through marine environments, others like Pierre Lermusiaux, Nam Pyo Suh Professor at MIT, use machine learning to better understand the marine environment itself.

Improving ocean modeling and predictions

Oceans are perhaps the best example of what is known as a complex dynamic system. Fluid dynamics, changing tides, weather patterns and climate change make the ocean an unpredictable environment that differs from one moment to the next. The ever-changing nature of the marine environment can make forecasts incredibly difficult.

Scientists have used dynamic system models to make predictions for marine environments, but as Lermusiaux explains, these models have their limitations.

“You can not account for every single water molecule in the ocean when developing models. The decision and accuracy of models and the ocean measurements are limited. There may be a model data point for every 100 meters, every kilometer, or, if you look at climate models of the global ocean, you may have a data point every 10 kilometers or so. It can have a big impact on the accuracy of your prediction, ”explains Lermusiaux.

Graduate students Abhinav Gupta and Lermusiaux have developed a new machine learning framework to help compensate for the lack of resolution or accuracy in these models. Their algorithm takes a simple model with low resolution and can fill in the gaps and mimic a more accurate, complex model with a high degree of resolution.

For the first time, Gupta and Lermusiaux’s framework learns and introduces time delays into existing approximate models to improve their predictive abilities.

“Things in the natural world do not happen instantly; however, all the prevailing models assume that things happen in real time, ”says Gupta. “To make an approximate model more accurate, the machine learning and data you enter into the equation must represent the effects of past states on the prediction of the future.”

The team’s “neural closure model,” which accounts for these delays, could potentially lead to improved predictions for things like a Loop Current vortex hitting an oil platform in the Gulf of Mexico, or the amount of phytoplankton in a given portion of the ocean.

As computer technologies such as Gupta and Lermusiaux’s neural closure model continue to improve and evolve, scientists can begin to unlock more of the ocean’s mysteries and develop solutions to the many challenges facing our oceans.

Leave a Reply

Your email address will not be published.