No abstract

No abstract

Tendons have been widely used for power transmission in the field of anthropomorphic manipulating systems. This article deals with the identification and enumeration of the kine matic structure of tendon-driven robotic mechanisms. The structural isomorphism of tendon-driven manipulators is defined, and the structural characteristics of such mechanical systems are described. Applying these structural characteris tics, a methodology for the enumeration of tendon-driven robotic mechanisms is developed. Mechanism structures with up to six degrees of freedom are enumerated.

The kinematic structure of tendon-driven robotic mechanisms has been investigated with the aid of graph theory. The correspondence between the graph representation of the kinematic structure and the mechanism has been established. We have shown that the kinematic structure of tendon-driven kinematic chains is similar to that of epicyclic gear trains. We also have shown that, using the concept of fundamental circuits, the displacement equations of tendon-driven robotic mechanisms can be systematically derived from the kinematic structure. The theory has been demonstrated by the kinematic analysis of three articulated robotic devices.

A method is proposed to improve the performance of the conventional G eneva mechanism. R ather than driving the input crank at a uniform speed, this method uses optimal control theory to synthesize the speed of the crank. In this paper, design criteria are rst developed based upon the elimination of the impact loading at the beginning and end of the motion cycle. In addition, three approaches to program the crank speed are introduced. In the rst approach, the crank speed is designed to be optimal so as to minimize the output acceleration. In the second approach, the crank speed is determined by minimizing the input motor torque. The third approach uses the degree of wear between the crank pin and wheel slot as the index to optimize the crank speed. All design objectives are formulated as a parameterized optimization problem and solved via an ef cient numerical method. F urthermore, trade-offs among the desired characteristics are taken into account by formulating the problem as a multiobjective optimization problem. Two examples are given to illustrate the design procedure.