Robots Imitate Life
The goal of robotics is to imitate what humans and animals make look easy.
Science Magazine published a special issue on “the social life of robots” (Oct. 10) that delves into technological and social issues of lifelike machines. Articles include:
- John Bohanon writes about the challenges of getting a robot to “see the big picture” the way humans do. “Robots are clumsy because they struggle to make sense of all the data coming in from their cameras,” he says. Keeping them from “floundering” in everyday environments is keeping researchers looking for the “next big advance.”
- Auke J. Ijspeert writes about the challenges of designing robots to move with the agility of animals. “The goals of biorobotics are to take inspiration from biological principles to design robots that match the agility of animals, and to use robots as scientific tools to investigate animal adaptive behavior.”
- Dennis Normile chuckles at how humans react to the presence of lifelike robots made by Hiroshi Ishiguro’s team. “In some situations, people prefer to speak with an android instead of another person, and they feel that robots should be held accountable for mistakes and treated fairly,” he says in “In Our Own Image.” He adds, “humans can quickly form deep emotional bonds with robots.” It’s not clear that the robots reciprocate.
- Adrian Cho writes about “the accidental roboticist” – evolutionist John Long’s attempts to get robots to evolve. To evolve what? Evolvability. That requires “modular design,” he is finding, showing that trying to understand a mindless, undirected process is requiring a lot of brain power.
The articles underscore the challenges of getting machines to act like people and animals, keeping human intelligent designers focused on the design around them. But Science Magazine is not alone; the imitation of nature in robotics and other technologies is popping up everywhere.
Meeting of minds: The IMDEA Institute is probing the interface “where humans, animals and robots meet.” Those three are all communicating complex systems. “In this context, a thriving technology trend applies the concept of animal swarms or swarming to the development of complex systems that bridge the gap amongst disciplines as dissimilar as biology, robotics or networking.” In trying to design “mixed swarms” able to solve distributed tasks, they’re getting swarmer.
Prosthetics: Advances in bionic technology would make Christopher Reeve proud, his son said for Live Science, a decade after the paralyzed actor died. “Electrical stimulation, brain-computer interfaces, exoskeletons and pharmaceutical therapies have proven somewhat effective at restoring mobility and other function to paralyzed individuals.” At the time, Reeve was campaigning primarily for embryonic stem cell research, but this article says nothing about it. The focus is entirely on bionics – technologies imitating nature.
Reach out and touch: Speaking of prosthetics, Science Magazine reported new work on sensor-equipped mechanical arms and limbs that are providing amputees with not just functional movement, but long-lost sensations of touch. These new devices are actually (see good photo on Medical Xpress) can actually give a patient the feeling that the prosthetic limb is part of his body. One patient said it removed the “phantom pain” that often plagues wearers of prosthetic devices. The advanced limbs are poised to improve the quality of life for amputees.
Microbots on the move: Visionary designer Robert Wood is featured in a National Geographic article about the swarms of microbots he is engineering. Some of them may work like programmable bees. One thing is clear: they don’t just happen. “Robots that fly. Robots you wear. Robots the size of nickels. These new classes of robots all have one thing in common—every aspect of them must be conceived and created from scratch.” If you want to make something small that flies, he says, there are a hundred thousand solutions all around you. “We try to understand the what, how, and why behind an organism’s anatomy, movement, and behavior, and then translate that into engineering terms.”
Sidewinder to the rescue: The media all got excited about Science Magazine‘s story of robotic snakes (see also John J. Chosa’s Perspective article in Science). Howie Choset at Carnegie Mellon had already succeeded getting a robotic snake to crawl, but it couldn’t climb up sand till he studied how sidewinders do it. PhysOrg and New Scientist, among other reporters, thought this was really cool as they posted the video clips of the mechanical sidewinder. Choset hopes that such devices can get into the nooks and crannies of collapsed buildings to look for survivors.
Teamwork: Speaking of search and rescue, the University of Pennsylvania’s ModLab thinks a combination strategy of snake and quadcopter would work best (PhysOrg). This solution takes inspiration from both snakes and birds. Video clips show the drone carrying the snake to a predetermined location, where the snake robot can crawl into a pipe or navigate through rubble.
Octopus bot: Watch PhysOrg‘s video of an octopus-inspired robot made by the Foundation for Research and Technology in Greece. Its tentacles pulse just like an octopus. The “Multi-Arm Robotic Swimming with Octopus-Inspired Compliant Web” was a hit at this year’s International Conference on Intelligent Robots and Systems in Chicago. “Now the researchers report that the addition of a soft and supple silicone web has nearly doubled the speed of the roboctopus, and not satisfied with that, the scientists have also taught it to crawl, carry objects, and swim free in the Aegean Sea,” an IEEE spokesman said. It didn’t even scare the fish; they apparently treated it as one of their own. Now they want to put a camera on it and see what else it can do.
Robotics is the ultimate biomimetics project. It’s also a profound example of intelligent design in applied science. If we design a perfect humanoid robot, after many years of research and development, would anyone be justified we succeeded by blind, unguided natural processes?