Tech Imitates Life: An Electric Fish-Inspired Robot Will Monitor Oil Spills, Ecosystems [VIDEO]

Black Ghost Knifefishen (Apteronotus albifrons en )

[Reporting from the annual Science/AAAS meeting in Chicago, Illinois]

Lacking proper fins and functional eyes, the electric knifefish (of the order Gymnotiformes; one of over a hundred freshwater species native to South America) has a remarkable ability to rapidly maneuver — and find its food — in the most cluttered and murkiest of waters. For this reason, the weakly electric knifefish has long intrigued scientist who knew of it, and, more recently, has helped inspire  new research on bio-mimetic (life-imitating) robots.

The over-arching goal of much robotics work these days is to achieve what’s known as intelligent autonomous robots…and what better models to use than those from the animal kingdom — be they cheetahs,  geckos, turtles, snakes, or insects like dragon flies…all such creatures have evolved special abilities that enable them to do remarkable things.

But it’s not just about imitation of animals (which is cool, of course, like MIT’s robotic cheetah); it’s about harnessing these remarkable abilities to go places where humans fear to tread, and, to do things that humans are simply unable to do.

In a symposium today entitled simply Electric Fish Robotics*, neuro-biologist turned robotic engineer Malcolm MacIver presented results of his lab’s prototype electric fish-inspired robot, and explained his motivation for doing so, as well as the electrosensory system behind it all.

“When people think of robots, they typically think of humanoid robots. These will have abilities that we already have, verses, abilities we do not have but which animals (like fish) do.”

The knifefish has what’s know as ribbon fins, similar to those of the oarfish, and are able to self-generate and electric current through fluid-filled compartment running along their bodies. The weak, milli-volt currents create a “traveling wave” effect which moves the ribbon fins in an undulating fashion (see the video, below) and allow it to swim. Simultaneously, the currents maintain an ambient (and very weak) electric field around the fish’s body which serves as the fish’s detection system. Also, the knifefish has over a hundred thousand electrical sensors embedded all over its skin. Thus, if an object in the fish’s environment has a stronger electric current (and, it will often have one if its a living creature) it will trigger the electro-sensors in that body location (i.e, the part of the body nearest the object). This all happens in a a fraction of a second, allowing the fish to maneuver with great rapidity through a completely dark, and complex, environment.

Thus MacIver calls the creature an “underwater acrobat”, and so did he choose to replicate this electric current and ambient field effect –both for sensing and moving — in his autonomous robot. Given a seemingly simple programmed goal such as find your way from one end of a tank to the other, the robot was able to successfully find its way around numerous columnar objects placed in its path — each time via a different route, as it currently has no memory function (but will need to have one eventually so as to identify target objects).

In robotics, there is typically  a trade-off between maneuverability and stability (i.e., the more maneuverable, the less stable), but by imitating the electric fish’s ability to propagate a current in two directions simultaneously — along  each of its ribbon fins (which is how it moves vertically, by the way) — MavIver has been able to overcome this trade-off. This represents a real advance in robotic engineering — especially for underwater robots.

Though still in the prototype stage, MacIver has big hopes for his robot, like assessing/monitoring underwater oil spills where the water is extremely murky, if not opaque, at times. There is also its unique, multi-directional propulsion technology which allows for underwater, “humming-bird like movement precision” that could come in very “handy” for recovery and repair missions and even conduct surveillance of threatened ecosystems like our planet’s quickly dwindling coral reefs.

And, with a name like MacIver, have no doubt the he will find many more uses for the robot. MacIver is currently a Professor of Biomedical Engineering at Northwestern University (Evanston, Illinois).

So then, here is a wonderfully compelling, if brief, video clip of the state of fish-inspired robotics, featuring the work of Malcolm MacIver and his lab 9article continues below):


To learn more about Malcolm A. MacIver’s work, you can download his papers Biomimetic and bio-inspired robotics in electric fish research, and, Mutually opposing forces during locomotion can eliminate the tradeoff between maneuverability and stability as a PDF files.

Top Photo: (Black Ghost Knifefishen – Apteronotus albifrons) credi: Derek Ramsey (Ram-Man); CC – By – SA 2.5



4 thoughts on “Tech Imitates Life: An Electric Fish-Inspired Robot Will Monitor Oil Spills, Ecosystems [VIDEO]”

  1. Consider Australia’s red dust (High iron content) that blows into the ocean every year , especially during droughts . Millions of tons per year

  2. It’s a really interesting concept, and I also remember being a bit aghast that these guys had gone off all loose-cannon style and just started dumping iron in the ocean. I’m curious as to what the potential downsides are, in a tangible sense. Obviously, the precautionary principle is usually a good thing, but it helps to know what the worst case scenario would be if something does go wrong. Iron can interfere with the food chain, cause potentially dangerous algal blooms, and the like. What else? Also…Is the Pacific Northwest a particularly good place to experiment?

  3. While Mr Whitehead seems genuinely interested in giving Russ George an objective shake, he [the author] is mislead or has simply overlooked a couple of major points.

    “However when water is warmed it tends to stratify, with a layer of
    warmer water forming on top of the cooler deep ocean water.
    Stratification of the ocean prevents mixing of surface and deep ocean
    waters, which results in surface waters becoming largely sterile.”

    Any oceanographer or sailor who has leaned over the gunwale far enough to get his hands wet reeling in the thermistor or sea anchor (respectively) knows that such stratifications are transitory and can be broken by waves and eddies driven by lasting winds. Destratification can come, as some might focus upon exclusively, before and/or after the peak of a plankton boom. In the real world, however, such bracketed synchronization between the bloom and destratification does not exist: in other words, slack winds do not always coincide exactly with blooms. Winds come up amidst blooms, obviously, and break up any stratification for several days or longer.

    In the real world, also, currents move through and just under the warm stratified layer (referred to sometimes as the Pycnocline), horizontally distributing bio-available nutrients along with senescent (dying) plankton far beyond the bloom area.

    Moreover, zoo-planketers migrate vertically (see diurnal or diel migration). There is great significance to this behavior – a significance totally overlooked by Mr. Georges less than fully informed critics. See Wikipedia entry for diel migration for this cite which well explains significance:
    “At night organisms are in the top 100 metres of the water column, but
    during the day they move down to between 800–1000 meters. If organisms
    were to defecate at the surface it would take the fecal pellets days to
    reach the depth that they reach in a matter of hours. Therefore by
    releasing fecal pellets at depth they have almost 1000 metres less to
    travel to get to the deep ocean.”

    Turning this post’s rhetoric on it’s head: there is no evidence to suggest “the
    experiment’ did not work and strong anecdotal evidence that it did.

    Moreover, the author jumps the shark on the notion that George’s approach was not adequately “scientific” – ignoring the confiscation of the data by a Federal SWAT team, just as field work was drawing to a close, overlooking the fact that the methodology for data collection and reduction has never been documented publicly (as the latter step could not be completed).

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top