Page 63 - Modern Robotics Building Versatile Macines
P. 63

LEARNING TO WALK   43




              CONNECTIONS: BEASTS AND BOTS

              Scientists and engineers since the time of Leonardo da Vinci (1452–
              1519) have studied biological anatomy and often tried to incorporate
              its principles in building such things as flying machines. Generally, it
              has turned out that direct imitation of nature’s mechanisms has not
              worked well with machines—thus airplanes do not fly the way birds
              do, although aspects of their wing structure are similar.
                In building robots (sometimes colloquially called “bots”) that have
              more of the capabilities of animals and people, today’s researchers
              have studied creatures ranging from the cockroach to dogs and even
              kangaroos. But at the same time that nature may be teaching the
              engineer, the engineer may also be helping to unlock the secrets of
              nature. As Raibert pointed out in his 1986 article, “Legged Robots,”
              in Communications of the ACM:


                   One way to learn more about plausible mechanisms for animal loco-
                motion is to build legged machines. To the extent that an animal and
                a machine perform similar locomotion tasks, their control systems and
                mechanical structures must solve similar problems. By building machines,
                we can gain new insights into these problems and learn about possible
                solutions.

                Another way to look at the possible convergence of biological
              and mechanical systems of control and locomotion is to consider
              how evolution has often guided different types of creatures to
              develop similar structures for similar tasks. (For example, insects,
              birds, and even mammals [bats] all developed wings for flight). In
              a way, robots, too, may be “evolving” similar structures to fit the
              niches or applications in which they will increasingly be found in
              the future.
                Just as Raibert found that relatively simple overall adjustments
              (such as changing stride length and posture) were enough to
              maintain balance while running, animal research is now suggest-
              ing that the nervous system does not “micromanage” locomotion
              but instead uses relatively simple overall controls. As Raibert told
              Stephen Budiansky, “the mechanical system has a mind of its
              own.”
   58   59   60   61   62   63   64   65   66   67   68