Mechanical Slithering: Coming to a Robot near You?
February 29, 2008
For his part, Dr. Barry Trimmer spends much of his time thinking about the caterpillar. Tufts’ Henry Bromfield Pearson Professor of Natural Sciences believes the insect can serve as a model for a new generation of robots made completely out of soft materials, enabling it to look, feel, and move in an organic, animalistic way, and even change shape.
“We’re trying to develop a type of robot that nobody’s ever seen before, with capabilities that no existing robot has,” he said. “And to do that, we’re using biology to help understand how we can control these types of robots.”
While conventional robots can be optimized for particular jobs, they lack the versatility of animals. Soft robots “will not be very efficient necessarily, but that doesn’t matter,” claims Dr. Trimmer. “If they can do a job that no other robot can do, it doesn’t matter if they’re efficient or not.”
Dr. Trimmer is the director of Tufts’ Biomimetic Devices Laboratory. Biomimetics, a relatively new field connecting science and engineering, is the study of the ingenious ways nature has developed to solve problems through evolutionary adaptation, their application to current technology. As biologist Leslie Orgel put it, “Evolution is cleverer than you are.”
Dr. Trimmer sees biomimetics not as a one-way street, but as a loop both increasing human beings’ understanding of the natural world and allowing them to reach new technical heights. Often, he said, his colleagues in the discipline fail to close that loop. “People take biology, and get inspired by it, and go and start to build something in an engineering context, but they then start solving engineering problems instead of going back and saying, ‘Wait a minute, what does this tell us about the animal?’”
Using biology to build robots also means using robots to study biology. Soft robots could be used to carry out experiments that would be impossible to conduct on animals. “We might discover that the robot is tolerant of damage, so if you remove a certain part of its controller, it still works,” explains Dr. Trimmer. “Well, we can’t ask a caterpillar to, you know, ‘Hey, stop using the muscle on the left.’” Despite these possibilities, Dr. Trimmer’s primary focus is to learn more about the systems for locomotion in the caterpillar so that he will be able to reproduce them in machines.
The task is more difficult than it sounds. Dr. Trimmer and his team, which began its work in earnest in 2005 and includes civil and environmental engineering professor Luis Dorfmann and environmental engineering professor David Kaplan as well as other faculty members, have faced challenges in trying to figure out how to control their soft robots, “because in the traditional way of thinking about control, you need to be able to plan where each part of the mechanism’s going to be.”
While the necessary calculations are relatively straightforward with the stiff materials and joints ordinarily used, they become much more complicated when one has to deal with soft, flexible materials. Dr. Trimmer, who received much of his training in neurobiology, hopes that his work on caterpillars will allow him to learn the principles of control embedded in their microscopic brains and apply them to their artificial relatives.
Building Robots for the Final Frontiers
Dr. Trimmer predicts that soft robots could have wide-ranging applications. Disaster situations, for example, often produce an extremely complicated debris field. Even today’s most versatile robots cannot hope to successfully navigate such areas, making them useless for relief efforts. “What we hope is that the robots will be able to change size and shape, and can also completely crumple up,” he said. “So these things could actually go into a very complex space, and perhaps have a camera or sensor, and find people.”He also thinks his robots, as a result of their kinship with insects, could be especially suited to help maintain and repair spacecraft. “Caterpillars are probably the most successful tree-climbing animals on the planet,” he said. “So they can climb in these very complicated three-dimensional structures really, really well. And if we are successful in mimicking what the caterpillars can do, we’ll have robots that can climb in three-dimensional structures.”
What does that have to do with space vehicles? “The antennae and the solar panels are absolutely enormous on [spacecraft],” he said. “Yet their main structures are about as thin as a drinking straw, so they’re extremely delicate.” Therefore, the work that astronauts can do on them is limited, but Dr. Trimmer’s machines could change all that. “What we’d imagine is that you’d have a squad of these little robots that are able to climb, and they would climb over the structure, find problems, maybe repair it,” he said.
He imagines that the robots could have remarkable uses in medicine as well. Just as caterpillars vary in size by a factor of 10,000 over their life cycle, but continue to use the same essential system with the same number of neurons and muscles at each stage, Dr. Trimmer thinks the technology behind soft robots, unlike other designs, could give rise to machines at a very large scale, a very small scale, and everywhere in between.
“So if that happens, you know, you can make these things go inside blood vessels,” he said. “Let’s say you need to find a tumor, and you’ve got a chemical detector that could actually find it, but you don’t know where to go, so you put it into the bloodstream, and when it detects, it sends out a signal that says, ‘I found it! It’s right here.’ That’s the sort of thing you could do.”
The Future Robotic
While some of these predictions may sound like science fiction, Dr. Trimmer expects to be able to unveil crawling robots this summer, and is now working on a two-year project to make a machine that can be deployed in the real world. “We already have prototypes that can move,” he said, “but they’re still pretty rudimentary.” Some of these prototypes are currently on display at New York’s Museum of Modern Art, featured in the exhibit “Design and the Elastic Mind.”Dr. Trimmer believes that soft robots in particular will become much more widespread in the coming decades. Already, he says, robots are in heavy use in industry, and he sees their role getting bigger and bigger, in part because of new manifestations like the ones he is developing.
Soft robots, like any technology, could have both beneficial and detrimental uses, Dr. Trimmer argued. “I mean, if you’ve got a robot that is silent, and is capable of changing its shape and squeezing under doors and down through keyholes,” he said, “that could be used for good or bad.” Ultimately, it depends on how society chooses to exploit them.
There are some social costs to Dr. Trimmer’s work, though, that cannot be so easily dismissed. Even in 2008, one can still find the odd young person at a party performing the series of contortions known as “the robot.” If Dr. Trimmer has his way, and soft, flexible machines begin to replace their jerky ancestors, these unfortunate souls will also be forced to learn a whole new set of moves.