On the Analysis and Kinematic Design of a Novel 2-DOF Translational Parallel Robot

The conceptual design of parallel robots can be dated back to the time when Gough established the basic principles of a device with a closed-loop kinematic structure (Gough 1956), that can generate specified position and orientation of a moving platform so as to test tire wear and tear. Based on this principle, Stewart designed a platform for use as an aircraft simulator in 1965 (Stewart 1965). In 1978, Hunt (1978) made a systematic study of robots with parallel kinematics, in which the planar 3-RPS (R-revolute joint, P-prismatic joint, and S-spherical joint) parallel robot is a typical one. Since then, parallel robots have been studied extensively by numerous researchers. The most studied parallel robots are with 6 DOFs. These parallel robots possess the advantages of high stiffness, low inertia, and large payload capacity. However, they suffer the problems of relatively small useful workspace and design difficulties (Merlet 2000). Furthermore, their direct kinematics possess a very difficult problem; however the same problem of parallel robots with 2 and 3 DOFs can be described in a closed form (Liu 2001). As is well known, there are three kinds of singularities in parallel robots (Gosselin and Angeles 1990). Moreover, not all singularities of a 6-DOF parallel robot can be found easily. But for a parallel robot with 2 and 3 DOFs, the singularities can always be identified readily. For such reasons, parallel robots with less than 6 DOFs, especially 2 and 3 DOFs, have increasingly attracted more and more researchers’ attention with respect to industrial applications (Tsai & Stamper 1996; Ceccarelli 1997; Tonshoff et al 1999; Siciliano 1999; Liu et al. 2001; Liu et al. 2005; Liu & Kim 2003). In these designs, parallel robots with three translational DOFs have been playing important roles in the industrial applications (Tsai & Stamper 1996; Clavel 1988; Hervé 1992; Kim & Tsai 2002; Zhao & Huang 2000; Carricato & Parenti-Castelli 2001; Kong & Gosselin 2002; Liu et al. 2003), especially, the DELTA robot (Clavel 1988), which is evident from the fact that the design of the DELTA robot is covered by a family of 36 patents (Bonev 2001). Tsai’s robot (Tsai & Stamper 1996), in which each of the three legs consists of a parallelogram, is the first design to solve the problem of UU chain. A 3-