The Scarlet Knight glider surfaces for pickup off the coast of Spain after its 5-month voyage across the Atlantic.

Has the age of human explorers come to an end? There are several projects underway that are sending robots where humans dare not tread. Scientists are ramping up to deploy hundreds of aquatic drones into the ocean depths, following the first successful transatlantic voyage of a robot last year. Swarms of miniaturized rovers the size of a penny will soon defuse mines, and thousands of terrain-ready robots will test rock composition on other planets. Machines will continue to explore where humans can’t, and swarm robotics is leading the way.

In 2009, Rutgers University launched an aquatic glider that became the first robot to cross an ocean. Nearly eight feet long, the device—named the Scarlet Knight, after Rutgers’ sports teams—spent months collecting data on the Atlantic. What makes the voyage remarkable is that the glider was at times directed from remote locations—once from as far away as Antarctica—using satellite, GPS and other technologies. Rutgers researchers breathed a sigh of relief with Scarlet’s success. Their first glider, launched the previous year, was presumably smashed by a shark before reaching the other side. Be sure to check out the video after the break.

Oceans cover nearly three quarters of the Earth’s surface, making fleets of data-collecting seabots invaluable to retrieving information about marine ecology, climate change, and the formation of hurricanes. Currently such data is gathered using satellites, buoys, and costly manned research vessels. Rutgers University oceanographer Scott Glenn says their new concept is all about omnipresence—getting a complete picture of what’s happening in our oceans. And to get a more complete picture, why not deploy more robots?

The National Science Foundation (NSF) agrees. The NSF recently gave $1 million to the Scripps Institute to fill the ocean with a fleet of drones that could potentially measure ocean drift, give detailed accounts of marine habitats, and collect other useful information. The Autonomous Underwater Explorer (AUE) project—led by the institute’s Jules Jaffe and Peter Franks—will eventually employ hundreds of robots that operate with a swarm mentality. There are many potential benefits to using swarm robotics. For instance, an individual drone like Scarlett can only collect data from its immediate vicinity, but a swarm can describe its relative movement to convey a larger picture. A swarm is also more resilient; judging by the fate of Scarlet’s predecessor, strength in numbers is a selling point indeed. Watch the NJN news segment below.

Once mass production begins on these new underwater drones, costs will drop. And at a significantly lower price tag compared to current methods, we may soon see their numbers grow exponentially. Even while Earth’s last great frontier increases in popularity, we haven’t yet turned away from exploring other planets. There are also a number of terrestrial bots in the works that will make poking around on land more efficient, too. Scaling down mechanical systems has made way for the Miniature Autonomous Robotic Vehicle (MARV)—one of the world’s smallest autonomous robots. Developed by Sandia National Laboratories, MARV was designed to complete difficult tasks like disabling mines and checking for chemical and biological weapons. An on-board computer can steer the vehicle and make other decisions, and at one cubic inch in size, that’s no small feat. With increased mobility and more intelligence, it’s only a matter of time before swarms of miniaturized rovers like these are sent on planetary missions.

At one cubic inch, MARV is one of the world's smallest autonomous robots.

The NASA Institute for Advanced Concepts (NIAC) supported research around a new approach to Solar System exploration that would deploy a large number of small spherical mobile robots (“microbots”) to moderate planets (including our own) and high temperature bodies, like Venus. Hundreds or thousands of robots would operate over rugged terrain while taking samples and measuring rock composition, allowing large-scale analysis to send home more details of what’s happening where no man has gone before.

Sending robots to do the dirty work of humans is central to the dogma of robotics research. But these advancements shouldn’t be seen as an end to the era of human exploration, but rather as the beginning of a fruitful collaboration between man and his creation. Just as humans are learning to cooperate more efficiently in numbers, we are teaching our robots to do the same. What remains to be seen is what exactly we may learn in return.

[image credit: Consortium for Ocean Leadership, ISRC]
[video credit: NJN Public Television]
[source: The Washington Post, ISRC]

The underwater drone swarm will eventually help researchers at UC San Diego explore the ocean (left). Right now, there are only five or six of the prototypes (right).

While the AUEs aren’t exactly articulated machines, they have many of the features and benefits of swarm robotics. As with many swarm robots, their strength is in numbers, and communication between individual bots. A solitary drone could only tell researchers about the conditions in its immediate vicinity. A fleet of drones will be able to describe their relative movement and the variations in ocean activity. It’s a cool concept that has great scalability. Right now Jaffe is planning on hundreds of drones, but imagine what we could learn with thousands or millions. The ocean is the last great frontier on Earth and these unmanned devices may be our best way of exploring it.

Jaffe is already building several prototypes for the AUE, including five or six of the soccer-ball sized ‘motherships’ and about 20 of a smaller version (the mini-AUE). The smaller and larger drones will be able to communicate via acoustic signals. A single mothership AUE could be moored while talking to free floating AUEs and mini-AUEs. The larger drones may have GPS trackers, while the smaller would not. This sort of network would allow Jaffe to measure ocean drift.

Other researchers have built buoyant drones before, some even more sophisticated. Yet none have taken measurements over the distance or period of time that Jaffe’s will. By dispersing a swarm of AUEs and mini-AUEs, Jaffe’s team could get a unique spatial and temporal understanding of the ocean.

Here's a better look at what's going on inside the AUE.

The following video is a presentation Jaffe gave explaining his AUE project. It’s a bit long and may not appeal to those unaccustomed to listening to scientific lectures. You can skip to 26:30 to hear Jaffe introduce the AUE device. He describes the mini-AUE (and shows a prototype) around 31:15. If you stick around to 38:25 and 47:30 he’ll tell you about the use of AUEs to locate plane crashes, and how building a mini-AUE might become a great project for middle school and high school students.

As Jaffe briefly described in the video, the AUEs could be modified with different kinds of instrumentation. They may measure salinity, dissolved oxygen, pH, chlorophyll levels, or turbidity of water. Each of these variables has a significant effect on ocean life. Taken collectively they could give a detailed account of marine habitats, helping governments decide if protected areas of the ocean are thriving.

But there’s really no limit to what these drones could learn. Biological info on marine habitats, meteorological on currents, seismological measurements on underwater earthquakes…the list goes on. Which is probably why the NSF has given Jaffe and his colleagues another $1.5 million to help with the control mechanisms for the movement of the AUEs. Each little drone is just a ball bobbing along under the water’s surface. As a collective they could be a vast net to catch all the secrets of the ocean. Maybe they’ll finally be able to tell me where the monster from Cloverfield came from.

[photo credit: Jaffe Lab, UCSD]
[video credit: UCTV]