mahdi tale masouleh

A Spherical Parallel Mechanism (SPM) is used to rotate a body around a fixed point. Using type synthesis, different kinematic arrangements can be obtained for the robot with three degrees of rotational freedom. The most commonly used structure for this robot is the 3-RRR kinematic architecture which is an overconstrained parallel mechanism and causes several problems of mounting the mechanism, but has the advantage of having high rigidity thus a good accuracy. In this paper two non-overconstrained architectures 3-RRS and 3-RSR that are extracted from the type synthesis are compared with overconstrained one from accuracy point of view based on joint clearance. First, a method to obtain a model of moving platform pose (position and orientation) error based on joint clearance is introduced which leads to a standard convex optimization problem. Then maximum values of six components of the pose error are computed in more than 1000 different configurations within their workspace. It is shown that this displacement is configuration dependent. The obtained results revealed that the 3-RRR SPM has better position accuracy while in the case of orientation, the 3-RRS SPM has lowest maximum error between SPMs under study in the prescribed workspace. It can be concluded that non-overconstrained structures can be used instead of the overconstrained structure. Finally, a comparison was made between the performance indices and the method presented in this paper and results showed that the kinematic sensitivity index is most consistent with the accuracy of the robot.

In this paper, a method is proposed which allows computing the position of the end-effector base on neural networks approach by considering external forces applied to the end-effector. As in under-constrained robots kinematics and statics are intrinsically coupled together, and they simultaneously should be considered, the forward kinematic problem of the robot converts to an optimization problem. Solving the optimization problem is time consuming and not suitable for practical purposes. So, in order to solve the forward kinematics problem a SimMechanics model base on the robots geometry and dynamic designed and presented. By means of this method, the forward kinematic problem is solved offline and used online. Moreover, an analysis of workspace is performed which reveals that the solution of the forward kinematic problem of the under-constrained cable robots can be calculated uniquely. By means of neural network method A position control is performed and the proposed method is validated. The comparison of operated and desired path is shown for a helical trajectory. Maximum error in the assumed workspace is 0.4 percent. Finally the proposed method was implemented experimentally and the results confirm the efficiency of the foregoing method.

This paper investigates the multi-objective optimization design of planar cable-driven parallel robots by using the evolutionary optimization algorithm. Since in cable-driven parallel robots، the cables should remain in tension in all configurations، the extent of the controllable workspace is considered as one of the design indices. This objective function is of utmost importance to the design of cable-driven parallel robots، since it considers the unidirectional properties of the cables in the analysis. In addition، in order for the robot to have suitable dexterity and accuracy and to be able to manipulate any arbitrary task in all the required directions، various kinematic indices such as global condition number، translational and rotational kinematic sensitivity indices are used. Through analysis of the conflict of these objectives within the workspace of the robot، it is shown that use of multi-objective optimization is an effective method to reach to a suitable trade-off. Furthermore، by applying multi-objective optimization methods such as the non-sorting genetic algorithm and the adaptive weighted particle swarm optimization algorithm، the optimal pareto front for the design parameters for the cable robot is obtained such that to draw a compromise between the robot designs.

This paper investigates the multi objective optimization of 6-degree of freedom cable-driven parallel robots by using the evolutionary optimization algorithm. In this regard, the determination of cable-driven parallel robots workspace is reviewed as the most important challenge in the design of space cable-driven parallel robots and among various definitions, controllable workspace is selected as a general definition of the cable-drivenparallel robots workspace, in which the robot cables remain in tension for any applied forces and wrenches to the end-effector. In order to evaluate the dexterity of the under study robot, the condition number index is used as an effective criterion to measure the distance from singularity. Moreover, the worst kinematic sensitivity is introduced as a presentable accuracy index. Furthermore, by taking the advantages of multi-objective optimization methods such as the non-sorting genetic algorithm, the optimal pareto front for the design parameters of the robot is obtained such that simultaneously, all of the robot design’s objectives are satisfied.