The distinctive whirring of stepper motors characterizes nearly all robots today, from fixed-in-place welders at General Motors to the mobile R2D2-style robots of science fiction. All that will change if professor Dennis Hong's whole-skin locomotion project is successful, since WSL robots will ooze toward their goals sans motors, gears or any conventional mechanisms.
Hong, a professor at the Virginia Polytechnic Institute and State University (Blacksburg, Va.), recently received a National Science Foundation Faculty Early Career Development Program Award for $400,000 to pursue WSL. NSF awards the five-year grant to creative junior faculty it considers likely to become academic leaders.
"WSL uses all of its surface for traction, which is why it's named 'whole skin locomotion,' " said Hong. "If successful, whole-skin locomotion would be the ideal locomotion strategy for search-and-rescue robots, as it will be able to squeeze under collapsed ceilings."
Success, however, is anything but assured for WSL. First of all, the everting motion of the outside skin, whereby its surface turns inside-out as it cycles around, begs for unconventional mechanical actuators that can expand and contract in rings. Such technologies exist, but they too are still in the experimental stages, making WSL doubly risky.
"We are not considering the conventional mechanical elements such as motors, gears and pulleys," said Hong. "For actuation methods such as using electroactive polymers, we will need more time." EAP, he said "is in itself an active research area that needs more advancement to become practical."
The second-biggest challenge is how to convert the contraction and expansion of unconventional actuators, such as EAP, into the cyclic everting motion of the outside surface. And related to that design problem is a looming packaging problem. Today, sensors stud the outside surface of mobile robots, enabling them to see and hear for purposes of obstacle avoidance and navigation. With the outside surface cycling around, however, there are no conventional locations to affix sensors.
"WSL is a new class of mechanism which converts the expanding and contracting motion of actuating rings into an everting motion," said Hong. "But the challenge not only includes the fabrication, but also addresses issues with packaging sensors and power systems, since the entire skin is moving."
Almost all mobile robots today have wheels, tracks or legs. Each of those mechanisms has its own strengths and weaknesses when it comes to the complex terrains rescue robots must traverse. To locate people trapped in a collapsed building, for instance, robots must be able to move over, under and between obstacles, as well as maneuver around tight corners. Likewise, a robotic endoscope has to bend around and squeeze through tight, uncharted curves.
"As the technology of robotics intelligence advances and new application areas for mobile robots increase, the need for alternative fundamental locomotion mechanisms for robots that can enable them to maneuver into complex, unstructured terrain becomes critical," said Hong. "Current methods of locomotion can do some part of this, but they have only had limited success in achieving all of these capabilities."
Hong has long been interested in harnessing the locomotion methods of single-cell organisms like amoebas, which navigate using either flagella, cilia or pseudopods. However, only the pseudopod--which extends the surface of the cell to ooze across, around and through objects--was suitable for ground-based robots. So Hong did a preliminary study on how to adapt the pseudopod to robotic locomotion, resulting in a design specification he dubbed whole-skin locomotion.
"To understand how amoebas move, we studied the biology of their motility mechanisms and applied their cytoplasmic streaming to a novel locomotion method, which we call whole-skin locomotion," said Hong.
To realize that design with available materials, Hong built an initial prototype using external quartz cords to drive a toroidal membrane. The moving membrane mimicked the everting motion of the endoplasm of an amoeba as it cycles around into the ectoplasm that motivates forward motion.
Now that Hong has received the NSF Career Award, he said he plans to spend the next five years creating a series of prototypes that not only demonstrate how to drive a toroidal membrane around a solid-tube core, but also resolve problems related to mechanics and packaging of WSL robots.
The first prototype, due early next year, will use electric motors inside a solid core to drive the cycling toroidal membrane. In the meantime, Hong's group will examine EAP and similar experimental technologies that can replace electric motors by configuring them as actuating rings that translate their expansion and contraction into forward motion. The EAPs will be placed at the head and tail end of the toroid to drive the membrane around the outer surface and propel the robot forward.
In 2008, Hong's group plans to build the first prototype using EAP actuators. After that, the group will address the problem of packaging the robot so that its sensors will remain stationary even when the outer skin moves.
Over the next five years, Hong hopes to build a body of scientific knowledge that others can capitalize on to craft their own prototypes using un- conventional locomotion methods, sensor mounts and actuators.
"Since our funding is from the National Science Foundation, our final goal is not to produce a prototype that can be manufactured, but rather to ad- vance the frontiers of science and technology so that we gain an understanding of what is possible for such novel locomotion methods," said Hong.
Hong is not banking all his bets on WSL, however. With funding from different sources, he is researching several other unusual locomotion methods at the Robotics & Mechanisms Laboratory (www.me.vt.edu/romela) at Virginia Tech.
"In my lab, we focus on research into many novel locomotion strategies," said Hong. "Besides the amoeba robot sponsored by the NSF Career Award, we are working on a wheel-leg hybrid robot, a three-legged robot and a humanoid robot, to name a few." Hong's three other innovative robot locomotion mechanisms include the Intelligent Mobility Platform with Active Spoke System, the Dynamic Anthropomorphic Robot with Intelligence and the Self-Excited Tripedal Dynamic Experimental Robot.
He advises Virginia Tech's Soccer Playing Robot with Intelligence, which competes each year in the RoboCup (www.robocup.org), an international autonomous-robot soccer competition. He is also one of the faculty members advising Virginia Tech's VictorTango team, which is entering the Defense Advanced Research Projects Agency's Urban Grand Challenge.
He is the recipient of a NASA Summer Faculty Fellowship and of the Young Investigator Award from the American Society of Mechanical Engineers Freudenstein/ General Motors. Hong also won the biomimicry best-paper award at the ASME Mechanisms and Robotics Conference.
Mobile robot mimics amoeba's everting motion with a toroidial membrane cycling around a solid-tube core.