Search

The avoidance of singularities is critical to design and control of parallel robots. This paper aims at efficient determination of the maximal singularity-free joint space and workspace of a class of six-DOF parallel robots with six kinematic chains of same type. We represent the singularity-free joint space by a 6-cube and determine it firstly. The singularity-free workspace is generated by continuous motion of all active joints in the singularity-free joint space. As a result, the boundary of the workspace can be obtained with simultaneous consideration of position and orientation of the mobile platform. The size relation between the maximal singularity-free joint space and workspace is discussed. To efficiently determine the singularity-free joint space and workspace, we propose dual quaternion-based Jacobian matrices and construct an efficient algorithm. The algorithm detects singularities in a given joint space and simultaneously calculates its corresponding workspace. The computational costs of the proposed algorithm and the traditional one are compared using a 6-U P S parallel robot, leading to 9 seconds and 458 seconds respectively. Finally, both the maximal singularity-free joint space and workspace of a 6- P US parallel robot are determined to further demonstrate the effectiveness of the new approach. [ABSTRACT FROM AUTHOR]

the six-axis vibration isolation system is essential for high-precision space systems. Its architecture is always designed based on the cubic-configuration Stewart platform, whose six flexible modes generally have different natural frequencies resulting in multiple resonances at various frequencies such that a uniform capability of vibration isolation cannot be achieved for the six flexible modes. To solve this problem, the dynamic isotropic design of the isolator is studied to make the six nonzero natural frequencies identical. The free-floating state of the isolator is taken into account and six design criteria of the dynamic isotropy are obtained. The internal relation between the dynamic isotropy and the kinematic isotropy is revealed and discussed. A decentralized active controller is then investigated for the isolator of dynamic isotropy. The controller decouples the six-axis vibration control into six identical control of a single-axis vibration isolator. The same control gains in each single-axis isolator reaches the optimum simultaneously for all the flexible modes such that a best performance of vibration isolation can be achieved. Finally, we present an example of an isolator of dynamic isotropy. With the proportional plus integral compensator, the uniform corner frequency and optimal active damping can eventually be achieved. [ABSTRACT FROM AUTHOR]

The forward kinematics is the basis of the design and control of the parallel robots. This paper aims to provide an efficient solution to the forward kinematics of a class of six-degrees-of-freedom parallel robots for real-time applications. With a unit dual quaternion used as the generalized coordinates of the robot system, the forward kinematic equations are derived to be a set of quadratic ones. An efficient algorithm is proposed to get the actual solution to them. The convergence and singularity problems of the new algorithm have been discussed. We have provided a convergence strategy and revealed the internal relation of the singularity with that of the parallel robot, proving the feasibility of the algorithm and giving the working condition in the practical applications. The new algorithm have been compared to the Newton's method for an 8-U P S parallel robot, resulting in the time consumptions of 0.2187 milliseconds and 14.25 milliseconds respectively. And then we perform a simulation of the state-feedback control for an 8- P US parallel robot. The two examples present the applications of the new algorithm and demonstrate its validity and efficiency. [ABSTRACT FROM AUTHOR]