In this study, a novel parallel learning particle swarm optimizer (PLPSO) is proposed. The evolutionary strategy of the algorithm is quite different from that of the existing PSO algorithms. To enhance the global using the PLPSO algorithm for solving the IK problem of a robotic manipulator is verified.

No abstract

No abstract

In this paper, a lightweight high-payload cable-driven serial-parallel manipulator is proposed. The manipulator comprises one 3-degree-of-freedom (3-DOF) shoulder joint and one single-DOF elbow joint. Using a special tension-amplifying principle, which is an ingenious two-stage deceleration method, the proposed manipulator has a higher load/mass ratio than those of conventional manipulators. In this paper, the special tension-amplifying principle is discussed in detail. The shoulder and elbow joints of the proposed manipulator are driven by cables. The design of this cable-driven mechanism and the mobility of the joints are analyzed, and the structural parameters of the joints are optimized to improve the payload capacity. The size of the manipulator is close to that of a human arm because the actuators of the cable-driven mechanism can be rear-mounted. Because the elbow joint is located at the end of the shoulder joint and the driven cables of the elbow joint are coupled with the rotation of the shoulder joint, the manipulator is designed with decoupled cable routing. The overall configuration and cable routing of the manipulator are presented, and then, kinematics, joint stiffness, strength, and workspace of the manipulator are analyzed. Finally, we report on the fabrication of a physical prototype and testing of its joint stiffness, payload capacity, workspace, speed, and repeatability to verify the feasibility of our proposed manipulator.

The large deployable mechanisms can be constructed by a set of basic deployable mechanical modules via mobile connection between the modules, i.e., the necessary mobility of each module is retained after connecting to other modules. In the large deployable mechanical networks, however, the mobility is always overconstrained due to the multiple close-loop structures in the mechanisms. This fact makes the mechanisms easily become rigid structures that are not movable even if small dimensional error happens to the mechanisms. Thus, eliminating redundant constraints so that the mechanisms still movable under the possible geometric dimensional errors, is very important for this kind of mechanisms. In this paper, efficient methods for the mobile connection between deployable single loop mechanisms (DSLMs) are first synthesized systematically; using some of these methods, lots of deployable units can be repeatedly used to build large deployable mechanical networks. A theoretic approach for eliminating the redundant constraints in the large deployable networks is proposed to make the deployable networks not so sensitive to dimensional errors. Finally, physical prototypes are fabricated to illustrate the feasibility of the proposed approach.