Page 111 - Human Inspired Dexterity in Robotic Manipulation
P. 111

Approaching Human Hand Dexterity Through Highly Biomimetic Design  109


              the two servos to bend and straighten the coupled fingers approximately
              once every 2 s. The coordinates of the reflective markers attached to the fin-
              gertip are recorded by a motion capture system (Vicon Bonita) composed of
              seven infrared cameras at 240 Hz VGA resolution. The Vicon system was
              calibrated, and is able to detect 0.5 mm displacement in all directions.
                 To avoid the marker occlusion and confusion issues, we attached a fore-
              arm frame (see Fig. 6.17) near the wrist of our robotic hand, so that we could
              record the marker trajectories for one fingertip at a time, and then transform
              their coordinates from the default world frame to this forearm frame when
              processing the data. The use of this forearm frame also allows us to constantly
              change the orientation of our robotic hand during hand motions to achieve
              good visibility for the reflective markers.
                 As shown in Fig. 6.18A and B, the Cartesian coordinates of the fingertips
              from the ring and little fingers are very similar to each other because their
              actuation is coupled (see Fig. 6.17). The data of 20 repetitions of the full
              flexion and extension motions show a highly repeatable pattern suggesting
              that our biomimetic finger design and the differential pulley system can be
              successfully combined to build a reliable robotic hand.
                 After being projected onto the X–Z plane, the flexion and extension tra-
              jectories of the ring finger’s tip can be clearly observed in Fig. 6.19B. Our
              experiments show that the flexion and extension motions of our biomimetic
              robotic finger were not following the same trajectory. The area bounded by
              the two trajectories covers a big portion of the reachable workspace of
              the ring finger [19]. The flexion trajectory closely resembles that of the log-
              arithmic spiral curve observed from a human finger’s flexion motion [20].



                300                            250
               Fingertip displacement (mm)  150 0  Fingertip displacement (mm)  150 0
                250
                                               200
                200
                                               100
                100
                                                50
                 50
                                               –50
                –50
                –100
                  0   10  20  30  40  50  60  70  –100 0  10  20  30  40  50  60  70
              (A)             Time (s)        (B)            Time (s)
              Fig. 6.18 The displacement of the fingertips during 20 repetitions of flexion and
              extension motions. (A) XY Z coordinates data from the ring finger. (B) XY Z
              coordinates data from the little finger. The tracking data were separately recorded
              for the ring and litter fingers to avoid marker occlusion and confusion. (A) The ring
              finger’s tip trajectory. (B) The little finger’s tip trajectory.
   106   107   108   109   110   111   112   113   114   115   116