SUMMARY Geometric accuracy is a critical performance factor for parallel robots, and regardless of error compensation, accuracy design or tolerance allocation is another way to ensure the pose accuracy of a robot at design stage. A general method of both geometric error modeling and accuracy design of lower-mobility parallel mechanisms is presented. First, a general approach for error modeling of lower-mobility parallel mechanism is proposed based on screw theory, and then the geometric errors affecting the compensatable and uncompensatable accuracy of the end-effector are separated using the properties of dual vector space. The pose error aroused by compensatable geometric errors can be compensated via kinematic calibration, while the uncompensatable geometric errors should be minimized during the manufacturing and assembly processes. Based on that, the tolerance allocation method is presented, giving each uncompensatable geometric error a proper tolerance by the use of reliability theory. Compared with the traditional tolerance allocation method, the advantages of the proposed method are as follows: the number of geometric errors to be allocated is greatly reduced; the results of serialized tolerance allocation can be obtained according to different reliability indices of pose accuracy of end-effector for designers to choose; on the premise of guaranteeing the same pose accuracy of end-effector, the allocated tolerances are loose and easy to realize. Finally, the proposed methods are successfully applied to an R(2-RPS&RP)&UPS lower-mobility parallel robot, and the effectiveness and practicability of the proposed method are verified.

No abstract

To investigate the effect of autogenous tissue engineering of growth plates in the treatment of growth plate injury. The growth plate chondrocytes were cultured from the iliac crest of 3-week-old rabbits by mechanical shearing and type II collagenase digestion. After in vitro development, the chondrocytes were seeded on the allogeneic decalcified bone matrix. After being mixed in culture for one week, the chondrocytes were implanted into the defects of the medial growth plate at the upper end of the right tibia; the left tibia was not treated. Dynamic X-ray photography was used to measure the shortening and angular changes in the lower extremity. The H & E and collagen1 immunohistochemical staining were used to observe the in vivo outcomes of the growth plate. There was a slight deformity in the right tibia of group A and group B on the 2nd and 3rd week after the operation, however, there was no significant difference between the three groups (P > 0.05). After that, the right tibia of group B and group C had progressive severe shortening and angulation deformity, while the right tibia of group A had no significant increase in deformity. There was a significant difference between group A, B and C at each time point (P < 0.05). In group A, the normal growth plate structure from collagen-1 immunohistochemical staining was recovered, while in group B and C the damaged area was repaired by new bone tissue. Autogenous tissue engineering of the growth plate can effectively prevent limb deformity after acute growth plate injury. The implanted tissue engineered growth plate can produce a columnar structure; cells can express type II collagen.