MIT's autonomous robotic tiles build floating structures on water

You don’t have to watch Kevin Costner’s Waterworld to know that for much of the world, the future will be increasingly flooded. As climate chaos causes oceans to swallow coastlands, and as surging water displaces and devastates communities, social survival will demand that people find ways to live and work with – and on – the water.

Fortunately, scientists at the Massachusetts Institute of Technology may have a partial solution.

Director of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) Daniela Rus, who is also the Panasonic Professor of Electrical Engineering and Computer Science at MIT, led the team that developed FloatForm. Floatform is a self-propelled, mostly-autonomous fleet of robotic tiles employing magnetic latches, sensors, and thrusters. These float-bots can assemble, detach, and reassemble themselves into platforms of varying shapes like Lego or the interchangeable world of the Micronauts.

In that future world beset by water, eventual developments with FloatForm could provide aquatic platforms for search-and-rescue, emergency medical service, shelter, remote maintenance, environmental monitoring, and even transportation, markets, schools, and performances.

Rus’s open-access paper in Nature Communications – with lead author Wei Wang and colleagues Niklas Hagemaan, Alejandro Gonzalez-Garcia, and Carlo Ratti – explores how FloatForm overcomes the numerous difficulties including complex nonlinear hydrodynamics and slipstreams from nearby robots to reconfigure without colliding. FloatForm units can even travel together using a hybrid coordination framework, while relying on minimal human guidance.

“With FloatForm, we are essentially turning static water surfaces into dynamic, programmable spaces,” says Wei, formerly of MIT and now head of the Marine Robotics Lab at the University of Wisconsin at Madison. He describes future cities in which “public space isn’t fixed, but can autonomously expand, contract, or reconfigure on demand.”

Currently, FloatForm is too small for that. Each robotic tile is about the surface area of a standard piece of paper (with the top folded down to make a square), 21 cm by 21 cm (8.27 in). However, co-author Ratti, the director of Senseable City Lab and professor of practice of urban technologies and planning at MIT, previously collaborated with the Amsterdam Institute for Advanced Metropolitan Solutions to deploy human-scale "Roboats" on Amsterdam’s canals.

Once focused on moving goods, those canals now focus on moving tourists. As co-author Hagemann says, the former project “explored whether the canals could be used for waste collection, or for transport, to offload some of the stress on the roads back onto the water. Hagemann, a CSAIL affiliate and MIT architecture graduate student who formerly researched at Senseable City Lab, investigated how to “expand public space onto water that’s currently underutilized” in increasingly dense cities.

“FloatForm projects envisions a future where the waterfront becomes a programmable extension of the city,” says Rus, with the tile robots proving valuable modeling for getting “dozens, and eventually thousands, of floating robots to organize themselves.” Such “distributed robotics opens new possibilities for mobility, emergency response, public space, and infrastructure on water,” she explains, including self-organized “bridges, platforms, and other useful structures on demand.”

Researcher Alejandro Gonzalez-Garcia, formerly of MIT CSAIL, expands upon Rus’ example. “If there’s an emergency, you could form a new bridge to alleviate traffic in the city, says Gonzalez-Garcia, who previously worked with the Senseable City Lab. “Or you could create floating markets and floating stages. If you want a more livable city, you want to use the water, too,” to form “infrastructure on the water, using a modular system to create one larger system.”

Inspiration for FloatForm arose from biomimicry.

In nature, thousands of autonomous fire ants using simple rules can create bridges and even rafts of their own living bodies. “We wanted each robot to have its own capabilities,” says Gonzalez-Garcia, “the same way ant colonies form a raft.” Such autonomy is mission-critical, because in a centrally coordinated system, if the coordinator can’t send signals or is destroyed, the entire coordination collapses, which in the real world could mean death.

But it doesn’t take a catastrophe to foil such a system – bandwidth alone is a major blocker. Without effective scaling, similar robots might assemble in a set order resulting in lengthy waiting of mostly inactive units. But FloatForm’s approach uses minimal central planning that gives each machine a final position in a larger assembly and delegating navigation, object-avoidance, and other adaptation to each robot responding to its neighbors so that, in current simulations, a fleet of 64 robots can move simultaneously.

“Every robot becomes an actuator,” says Gonzalez-Garcia. As Hagemann explains, FloatForm units “become more stable by joining together, like the ant raft, if you have waves or currents.” Furthermore, while the robots are too small to hold large batteries, they use minimal power for thrust, and employ magnets to grasp nearby units up to 15 cm (5.9 in) away. “If they use less energy on latching,” says Gonzalez-Garcia, “they can use more on computation, or on actually moving.”

However, adapting FloatForm to the human-scale world won’t be simple. “These boats are very small, so in very disturbed water, they cannot work,” says Gonzalez-Garcia, given that a ripple on the surface of a tank is nowhere near the threat of river slipstreams or ocean waves. Human-scale FloatForms will require reinforced and even mechanically interlocked latches as well as GPS or visual sensing.

But once the team solves those problems, deployment could be nearly anywhere with bodies of water. “Venice, the Netherlands, Belgium, the fjords and lakes of Norway, really any city with a river can take advantage of this,” says Gonzalez-Garcia. “Where else can water be used for something more?”

Source: MIT