Biologists and engineers have joined forces to build a new robot bat that’s helping us understand how real bats use echolocation to hunt for food. By creating a robot that can echolocate, the team mimicked a bat’s flight path and explained how bats can quickly determine whether or not their prey is on a leaf. This new bat’s eye view is detailed in a study recently published in the Journal of Experimental Biology The study was led in part by bat scientist and Smithsonian Tropical Research Institute research associate Inga Geipel. In fact, the robot’s performance largely confirmed Geipel’s hypothesis about real bats. While she expected these results, she still found them gratifying, not so much for herself, but for her furry subjects.  “I’m always Team Bat,” Geipel tells Popular Science. “They always trick me, they always outsmart me.”  Echolocation for people in a hurry Bats use echolocation to find their way and hunt for prey. The winged mammals emit rapid clicking sounds from their mouths and listen for the echoes as those sounds bounce off nearby objects, which could include potential meals. By interpreting the timing and strength of those returning echoes, bats can build a detailed acoustic picture of their surroundings.  That sonar-based perception process is somewhat akin to the way autonomous vehicles use LiDAR sensors to create a mini map of the world around them. But while self-driving cars rely on dozens of cameras and sensors working in concert, bats accomplish the same task intuitively, with just two ears and a mouth. Though scientists have long known that bats used echolocation, it was still unclear exactly how they utilize it in the real world, especially in densely packed jungles and rainforests where there are virtually unlimited empty leaves vying for a hungry bat’s attention.  Common big-eared bat (Micronycteris microtis) eating a freshly-caught dragonfly. Image:Christian Ziegler. Entering a bat’s world  To fly into a bat’s world, the team built on Geipel’s nearly 20 years of research. She says her fascination began when she glimpsed one of the flying mammals deftly fluttering through a lightless night sky. Also a lifelong admirer of music and sound, Geipel was captivated by the notion that these creatures could use those senses to “see” in ways humans can’t comprehend. She hoped her future work would shed some light on that intellectual darkness. “Seeing the world through sound is a sensory system that is alien to us,” Geipel said. “I find it highly fascinating that bats can fly in total darkness.” The new robot bat study is something of a spiritual “sequel” to Geipel’s PhD research on bat foraging. That earlier work showed that big-eared bats (Micronycteris microtis) initially approach leaves at a specific angle so that their sonar clicks reflect off smooth forest leaves like an echolocation mirror. Leaves with objects on them, such as  insects, scatter the sonar, resulting in the bat receiving a stronger return pulse. From the bat’s perspective, stronger echoes can mean a tasty lunch. Related Stories World’s largest carnivorous bats are big softies A furry antelope robot is keeping tabs on its organic cousins LASSIE’s robot dog may join astronauts on Mars How bats avoid crashing into one another But while that basic theory makes intuitive sense, it also presents a practical problem. For the proposed system to work, bats would seemingly need to know the orientation and position of every leaf they pass, whether or not it holds potential prey. In a forest, a hungry bat would be overwhelmed by the need to constantly analyze a cacophony of echoes from countless leaves muddying its sonar. The bat would essentially spend all its time toiling over the correct angle of approach. “Behavioral experiments had already suggested how these bats might solve the problem of finding prey-occupied leaves, but we wanted to know whether that explanation was actually sufficient to make the behavior work,” paper co-author and University of Cincinnati associate professor of biology, mechanical engineering, and electrical engineering  Dieter Vanderelst said in a statement.  That’s where the idea for the robot came into play. The robo-bat was designed to function as a mechanical stand-in for the real thing, allowing researchers to analyze how the winged mammals approach leaves with and without prey. To do that, the team brought together experts from both biology and engineering in pursuit of a common goal—an interdisciplinary collaboration that isn’t all that common. Geipel says the team drew on the combined knowledge of biologists like herself and engineers capable of modeling the physical world through robotics. “By building the bat’s hypothesized foraging strategy into a robot and testing it in the physical world, we could ask whether a simple, elegant solution can succeed under complex acoustic conditions, ” Vanderelst added. Designing a bat robot When designing the robot, the team wanted a system that closely modeled a bat’s natural foraging technique without adding unnecessary complexity. The resulting bat robot” prioritizes function over form. It consists of a robotic arm with a built-in sonar emitter meant to mimic the chirps a bat produces. At the end of the arm are binaural microphones that serve as the robot bat’s “ears.”  The entire apparatus is mounted on a 9.8-foot -long (or three meters) linear track, which functions as a highly condensed flight path. The track is so condensed that it  fits in what looks like a small office. Common big-eared bat (Micronycteris microtis) approaching a katydid resting on a leaf. Image: Inga Geipel, Smithsonian Tropical Research Institute. The robot performs the tasks necessary to collect crucial research data, but it certainly wouldn’t fool anyone into thinking it was a true doppelganger of its biological inspiration. Personally, Geipel says she would have favored adding googly eyes, but they ultimately passed on the idea for the sake of professionalism. The “leaves” in this case were 3D-printed cardboard. Some of them had a roughly 3.5-inch-long (nine-centimeters) 3D-printed cardboard dragonfly pinned to their centers to represent potential prey. During the experiment, the robot moved along the track, emitting successive sonar pulses with about a 0.5-second delay between them. The resulting signal data formed what the researchers call an “echo envelope,” which was then wirelessly sent back to the computer controlling the robot arm. Robotic arm equipped with a sonar head searching and finding an artificial dragonfly on artificial leaves. The laser indicates where the SONAR head is looking. CREDIT: Dieter Vanderelst, University of Cincinnati In total, the team conducted more than 45 trials of the robo-bat flying past various leaf configurations, both with and without prey. The system performed remarkably well. The robot successfully detected leaves with a pinned dragonfly 98 percent of the time and falsely identified prey on empty leaves only 18 percent of the time.  Critically, the bat robot achieved these results without first assessing the orientation or angle of the leaves, one of the primary questions the researchers aimed to answer. The  bat appeared to follow a simple framework: track strong, stable echoes above a certain threshold and ignore those that don’t meet it.  While this work specially looks at big-eared bats (Micronycteris microtis), the researchers are hopeful they could apply it to other species. Bat inspired robots come in several shapes and sizes   Bats have inspired other robots in past studies.  In 2017, engineers at Tel Aviv University developed Robat the Robot, a first-of-its-kind autonomous, wheeled device that navigated and explored its surroundings solely using echolocation. Although it couldn’t fly, Robat was equipped with an ultrasonic speaker that emitted bat-like chirps every 30 seconds. It processed the returning echoes through an onboard machine-learning model, which allowed it to identify and avoid obstacles in real time. Before that, researchers from Caltech and the University of Illinois at Urbana–Champaign designed Bat Bot, a bat-inspired robot with soft, articulating wings that weighed just 3.2 ounces (93 grams). The major innovation there was the creation of synthetic wings capable of changing shape as they flap, much like those of a real bat. The team achieved this by developing a custom-made, ultra-thin silicone membrane for the wings. The robot Geipel and her colleagues helped create, by contrast, might be less visually impressive than these two earlier examples. However,  its function arguably provides researchers with richer data to actually understand with better detail how real living and breathing bats operate.  Looking ahead, Geipel says she and her team hope to expand the research to include a wider range of bat species and see if they can understand more clearly how bats distinguish between different kinds of possible prey clinging to leaves. When it comes to studying bats more broadly, she adds, there is still plenty left to uncover. “We are just scratching the surface here,” Geipel said.  The post A robot bat sheds new light on how they hunt in darkness appeared first on Popular Science. ...read more read less