Octopus-Inspired Robot Matches Real Octopus For Speed

Underwater vehicles have never matched the extraordinary agility of marine creatures…until now

Humans, and the machines we make, are particularly unwieldy swimmers. Many types of fish can travel at speeds of up to 10 body lengths per second. By contrast, an Olympic swimmer manages no more than about one body length per second. And while nuclear submarines can travel at about 80 kilometres per hour, that translates into much less than half a body length per second.

This kind of performance has long fascinated marine engineers who would dearly love to match it. In particular, they have puzzled over the way certain sea creatures can accelerate from a standing start at rates of up to 120 metres per second squared. That’s 10g!

Among the creatures that demonstrate this extraordinary fast-start acceleration are the octopus, squid and certain jellyfish. Marine biologists have long observed these creatures’ ability to avoid prey by filling their bodies with water and squirting it out to generate propulsion. In this way, the octopus, for example, can accelerate at more than 10 body lengths per second squared.

That gave Gabriel Weymouth at the University of Southampton and few pals an idea. Why not build their own robotic octopus capable of squirting in the same way and seeing whether it can match its natural cousins’ performance. The results are something of a surprise.

Their underwater robot is essentially a copy of an octopus’s head, which expands when it is filled with water and squirts it out through a nozzle at the rear. This robot has no tentacles but instead, the team fitted small fins to the rear to stabilise its movement through the water.

So the robot is roughly ellipsoidal in shape, measuring 27 centimetres in length and is 5 times longer than it is wide. It is created by stretching a rubber membrane over a set of polycarbonate ribs. Weymouth and co chose this shape because of its ability to glide many body lengths through water.

The team then filled the robot with water under pressure, causing the rubber membrane to expand into a bluff body shape that is particularly unsuited to gliding. Finally, they released the robot in a pool and filmed its progress through the water at a rate of 150 frames per second as water squirting through a rear-facing nozzle propelled it along.

The results are extraordinary. As the water starts to squirt out of the nozzle, the robot begins to accelerate. However, it moves little during the first half second or so since its bluff body shape prevents efficient progress.

But then, as the robot shrinks, it accelerates rapidly. And here’s the thing: Weymouth and co say that its maximum acceleration is 14 body lengths per second squared and that it reaches speeds of 10 body lengths per second.

That’s unprecedented for an underwater robot. “The robot is found to experience extraordinary speeds,” say Weymouth and co.

What’s more, the peak force on the robot is 30 per cent greater than the thrust created by the jet. By contrast, a robot that expels water without shrinking never experiences a force greater than the thrust from its jet.

That raises the obvious question of how this happens. And Weymouth and co think they have the answer.

One of the key properties of efficient movement through water is that the flow around a body must be smooth or laminar. That’s why streamlined shapes are so important.

However, when this smooth flow separates from the body, the result is turbulence, the great enemy of all hydrodynamicists that increases drag dramatically.

Bluff bodies that are not streamlined tend to generate just this kind of turbulent flow, which is why they do not move through water very efficiently.

Aerodynamicists have tried all kinds of tricks to prevent the separation of laminar flow. One of the most exotic is to use membranes covered with tiny holes that allow water to pass through.

If the water is sucked through these holes as the structure moves through the water, the laminar flow remains attached to the surface. This reduces drag and dramatically improved the movement through the water.

In practice, however, this tends to be a complicated approach that has been hard to perfect.

But Weymouth and co say a shrinking octopus seems to work in just the same way. As the membrane shrinks, it pulls at the surrounding water, ensuring that the laminar flow stays attached to the surface as it shrinks. And that prevents the formation of turbulence.

This sucking effect also injects energy into the water flow that had previously been stored in the stretched membrane. That’s why the peak thrust is higher than the force produced by the jet.

Weymouth and co say the effect should scale with the size of the robot, provided stiffer membranes can be developed to store energy. “For a given robot design and higher available energy, the efficiency is expected to remain near the (high) level found in these experiments, and an increase in membrane stiffness will enable man-made vehicles to rival the acceleration of their biological inspirations,” they conclude. Indeed, they say the fast-starting performance should improve with increasing size.

That’s fascinating work that should lead to a new generation of underwater vehicles that can move at unprecedented rates.

That should interest the military — an obvious application is in underwater missiles. But marine biologists might also be interested in building artificial sea creatures that can match the behaviour of their biological peers. It might even provide inspiration for engineers entering the annual human-powered submarined contest, the next of which takes place in June next year.

Ref: http://arxiv.org/abs/1409.3984 : Ultra-Fast Escape Maneuver Of An Octopus-Inspired Robot