Robots Are Helping Us Understand Nature

January 18, 2022 steven

More and more, scientists are turning to robots to collect the data they need to conduct research regarding natural habitats and the animals that reside in them.

Wouldn’t it be great to have your own robot; a mechanical butler whose sole purpose is to make your life a bit easier by bringing you things and doing your cleaning and cooking? After all, books, movies and plays set in the near future usually show us such a world and create our yearning for it.

In truth, robot technologies that seamlessly mesh with our personal lives will likely happen within the next 20 years. That has me thinking: how will robots soon influence our interactions with and studies about the natural world?

I don’t have to wonder all that hard. Robots are already doing some of the heavy lifting when it comes to understanding nature—its wild places and wild inhabitants.

The Galapagos Islands are considered one of the world’s foremost destinations for wildlife-viewing. Isolated terrain shelters a diversity of animal and plant species, many found nowhere else. ©Pedro Szekely, flickr

Learning from robotic Galapagos lizards

The United Nations Educational, Scientific and Cultural Organization (UNESCO) describes the 19 Galapagos Islands and the surrounding marine reserve—a World Heritage site—as “a unique ‘living museum’ and showcase of evolution. Located at the confluence of three ocean currents, the Galapagos are a melting pot of marine species. Ongoing seismic and volcanic activity reflect the processes that formed the islands. These processes, together with the extreme isolation of the islands, led to the development of unusual animal life—such as the land iguana, the giant tortoise and the many types of finches—that inspired Charles Darwin’s theory of evolution by natural selection following his visit in 1835.”

And it’s here, in this one-of-a-kind archipelago, that scientists first used a robot that interacts with wild subjects in real time to advance knowledge of a wildlife species.

To avoid injury from male-to-male contests, some animals display behaviors—such as color changes or sequences of movements—that showcase body size and fighting ability. In Galapagos lava lizards, one of the most recognized behaviors is a head bobbing (or push-up) display.

Lava lizards are some of the most abundant reptiles throughout the Galapagos. They’re often found in large groups warming up in the sun atop lava, which gives them their common name. ©Steven dosRemedios, flickr

Seeking to learn whether the lizards would react more quickly and strongly to their opponent’s bobbing display if it occurred immediately—as opposed to if there was a delay following the initial challenge—scientists used remote-controlled, lizard robots made from hand-carved wood, latex limbs and high-resolution photos.

The researchers positioned the robots approximately three to 10 feet from 20 wild Galapagos lava lizards found on the island of San Cristobal. After provoking an initial response by the native lizard, the researchers remotely activated the robot lizard to respond with a preset countermovement, either immediately or after a 30-second delay.

What they found was that male lava lizards are sensitive to the timing of their opponents’ responses during contest displays, with quicker responses being perceived as more aggressive. An immediate response by the robot stimulated the wild lizard to react more significantly and quickly more often than when the robot’s response was delayed by 30 seconds.

Like all Galapagos species, lava lizards face the threats of El Nino events, global warming and habitat destruction caused by humans.

The authors of this investigation suggest that the live lizards’ abilities to assess their contestants’ levels of aggression may help the animals size up their competitors, influencing their decisions to either retreat or to instigate a battle, helping them avoid disadvantageous injuries.

Previous research in this area has used either prerecorded video playback or robots with movements set on a loop. These findings, however, confirm that realistic robotic stimuli can be used to interact with wild animals, to communicate with them and even manipulate their behaviors. They also further our understanding of how lava lizards communicate with each other in their natural habitats.

Enlightenment from robotic Antarctica observers

Robots are not only helping us understand the natural world at the equator, but at the poles, as well.

The difference between sea ice and ice shelves is that sea ice is free-floating, whereas ice shelves are firmly attached to the land. Sea ice contains thin sea ice; thicker, multiyear sea ice; and icebergs.

The Ross Ice Shelf, a part of the Antarctic Ice Sheet that is floating on the ocean, is currently estimated to cover an area of 182,000 square miles, making it roughly the size of the Yukon Territory in Canada. The shelf’s mean ice thickness is about 1,100 feet. Its magnitude—and the fact that thinning of the ice shelf will speed up the flow of Antarctica’s ice sheets into the ocean—means that it carries significant sea-level rise potential if it were to melt. Melting ice shelves like the Ross could cause seas to rise by several feet over the next few centuries.

Traditionally, data on ocean circulation, depth, salinity and temperature around the ice shelf is obtained in two ways: deep moorings and research cruises. Because the Ross Sea is covered by sea ice for most of the year, ship-based measurements are restricted to a short period in the high austral summer. Moored sensors, on the other hand, can collect data for several years; however, they are generally deployed no higher than about 650 feet below the water’s surface in order to avoid passing icebergs, so they provide a less complete picture of what’s happening around the ice shelf.

But a new approach that uses robots to gather data from the Ross Sea offers fresh insights into the forces causing the world’s largest ice shelf to melt. In a study published in the Journal of Geophysical Research: Oceans, researchers deployed six robotic floats, called Air-Launched Autonomous Micro Observers (or ALAMOs). They fastened parachutes to the floats and launched them out of a New York Air National Guard airplane from 2,500 feet above the icy waters. The robots were programmed to avoid sea ice that could damage their external sensors and antennae. The floats were then “parked” on the seafloor between data gathering sessions to limit their drifting on ocean currents.

In Antarctica, ALAMO robots, deployed by parachutes, were programmed to avoid the sea ice that could damage their antennae and external sensors.

The floats gathered salinity and temperature readings from the seabed to the surface, sending back data to a research team by satellite every day. Seven other floats, deployed from a ship three years earlier, provided records of ocean conditions farther north, away from the ice shelf.

In other places in Antarctica, ice shelves are being melted by flows of global warm water from the deep ocean to the coast. But the researchers found that local factors are influencing the Ross Ice Shelf’s stability, refining predictions of how it will change and influence sea rise in the future.

For the Ross Ice Shelf, the main source of ocean heat that’s causing it to melt is sunlight warming the upper ocean after the region’s sea ice disappears in summertime. Sea ice normally reflects sunlight, whereas darker seawater absorbs it. The research team also measured large amounts of fresh water coming into the Ross Sea from rapidly melting ice shelves in the Amundsen Sea to the east. Once this extra fresh water reaches the ice front, it changes how heat mixes down from the surface to the base of the ice shelf, where melting occurs, meaning that future Ross Ice Shelf stability depends on changing coastal conditions in both the Amundsen Sea and close to the ice shelf front.

An international team of penguin experts concluded that 70 percent of emperor penguin colonies in Antarctica could become extinct by 2050 if the current rate of sea-ice loss continues.

The scientists noted that increased ocean heating and ice-shelf melting could occur if the summer season, during which the sea is free of ice, becomes longer.

This novel approach to collecting local data from remote Antarctica’s continental shelves provides a new way to check the reliability of the global, numerical models and will be critical in narrowing the range of predictions regarding how much ice Antarctica will lose in future climates and how high seas will rise.

Inspiration from cheetah-like, robotic leapers

Cheetahs are the fastest creatures on land, and they derive their speed and power from the flexing of their spines. Inspired by the biomechanics of cheetahs, National Science Foundation-funded researchers have recently developed a new type of soft robot capable of moving more quickly on solid surfaces or in the water than previous generations of the robots.

In a paper recently published in the journal ScienceAdvances, scientists state that they’ve built a type of soft robot based on a cheetah’s movements that has a spring-powered, “bistable” spine, which means that the robot has two stable states. It’s possible to rapidly switch between these two conditions by pumping air into channels that line the soft, silicone robot. This releases a significant amount of energy, allowing the robot to quickly exert force against the ground, enabling it to gallop (with feet leaving the ground) across the surface.

Cheetahs are built for grace, hunting and speed. They’re the world’s fastest land animals, capable of running up to 70 miles per hour.

Until now, the fastest soft robots could move at speeds of up to 0.8 body lengths per second on flat, solid surfaces. This new class of soft robots, called Leveraging Elastic Instabilities for Amplified Performance (LEAP), can reach speeds of up to 2.7 body lengths per second—more than three times faster. The new robots are also capable of running up steep inclines, which can be challenging or even impossible for soft robots that exert less force against the ground.

The researchers report that they are optimistic that they can modify this design to make LEAP machines that are even faster and more powerful, paving the way for robots that, like humans, will be able to perform multiple functions, such as running, swimming, and grabbing and lifting objects.

Potential applications include search-and-rescue technologies, where speed is essential; caregiving; and industrial manufacturing, with production-line robots that are faster but still capable of handling fragile objects.