No abstract

SUMMARYThis paper describes a numerical algorithm to solve the inverse kinematics of parallel robots based on numerical integration. Inverse kinematics algorithms based on numerical integration involve the drift phenomena of the solution; as a consequence, errors are generated when the end-effector location differs from that desired. The proposed algorithm associates a novel method to describe the differential kinematics with a simple numerical integration method. The methodology is presented in this paper and its exponential stability is proved. A numerical example and a real application are presented to outline its advantages.

SUMMARYPick-and-place applications need to perform rigid body displacements that combine translations along three independent directions and rotations around one fixed direction (Schoenflies motions). Such displacements constitute a four-dimensional (4-D) subgroup (Schoenflies subgroup) of the 6-D displacement group. The four-degrees of freedom (dof) manipulators whose end effector performs only Schoenflies motions are named Schoenflies-motion generators (SMGs). The most known SMGs are the serial robots named SCARA. In the literature, parallel manipulators (PMs) have also been proposed as SMGs. Here, a novel single-loop SMG of type 2PRPU is studied. Its position analysis, singularity loci and workspace are addressed to provide simple analytic and geometric tools that are useful for the design. The proposed single-loop SMG is not overconstrained, its actuators are on or near the base and its end effector can perform a complete rotation. These features solve the main drawbacks that parallel SMG architectures have in general and make the proposed SMG a valid design alternative.

Manufacturing and assembly (geometric) errors affect the positioning precision of manipulators. In six degrees-of-freedom (6DOF) manipulators, geometric error effects can be compensated through suitable calibration procedures. This, in general, is not possible in lower-mobility manipulators. Thus, methods that evaluate such effects must be implemented at the design stage to determine both which workspace region is less affected by these errors and which dimensional tolerances must be assigned to match given positioning-precision requirements. In the literature, such evaluations are mainly tailored on particular architectures, and the proposed techniques are difficult to extend. Here, a general discussion on how to take into account geometric error effects is presented together with a general method to solve this design problem. The proposed method can be applied to any nonoverconstrained architecture. Eventually, as a case study, the method is applied to the analysis of the geometric error effects of the translational parallel manipulator (TPM) Triflex-II

The dimensional synthesis of translational parallel manipulators (TPMs) of type PRRR-PRPU is addressed by using an overall novel method. Addressing this design step on such TPMs is interesting for the scientific community since, in a previous paper, one of the authors showed that it has the following promising features: a single-loop not-overconstrained architecture with all the actuators on or near to the base, a simple position analysis, easy-to-find workspace boundaries, no constraint singularity, a type-II singularity locus that is a plane easy to keep far from the useful workspace, and a double infinity of isotropic configurations. The presented synthesis includes the analysis of isotropy and dexterity by local and global conditioning indexes, the useful workspace optimization, and the accuracy and stiffness analyses. The result is the identification of a normalized TPM of type PRRR-PRPU with performances that are comparable with those of commercial TPMs.The identified normalized TPM yields a set of actual TPMs with the same performances by changing the value of a reference geometric length. Also, the chosen shape of the useful workspace (i.e., a cuboid) matches the needs of many industrial applications.

The growing interest in use of renewable energy sources, such as photovoltaic energy systems, occurs due to the high cost of conventional energy sources and the environmental awareness linked to renewable sources. For photovoltaic panels efficient operation, it is necessary the system presents appropriate cleaning conditions to the dirt do not obstruct the solar radiation incidence. In this context, periodic cleaning of photovoltaic panels is an obvious necessity. This work aims to present a market survey and patent analysis on the use of robots to perform cleaning tasks on photovoltaic panels. For that, the Brazilian and international literature were consulted. As a result, it was noted the existence of different solutions for cleaning photovoltaic panels, all with positive and negative aspects in practical terms. With this study it was also possible to map the technology of robotics for cleaning photovoltaic panels.

No abstract

parallel robot imposes constraints on the moving platform to apply the degrees-of-freedom specified in the design of the robot. Without a change in the design of the robot, it is not possible to change its geometry before or during the execution of a task. Due to the closed-loop kinematic chain, in general, parallel robots require high manufacturing precision because any misalignment in the joint axes can cause internal overconstraints which are harmful to the robot.A characteristic of the new class of variable-configuration parallel manipulator proposed in this paper is that its form can be changed without changing the degrees-of-freedom and the characteristics of the motion of the moving platform. The change in the configuration of this new class of parallel manipulator is managed by additional degreesof-freedom called degrees-of-freedom of self-aligning. These additional degrees-of-freedom of self-aligning are passive/nulls, they permit a change in the configuration before starting the task execution but during the task execution they do not have any influence, i.e. their velocities are nulls.As an initial study on this new class of parallel manipulators, we address a change in the geometry of the base of the parallel manipulator. The geometry of the base of this parallel manipulator can change completely depending on the shape of the floor and this change is managed by degrees-of-freedom of self-aligning. These degrees-of-freedom of self-aligning give additional degrees-of-freedom to the kinematic chain of the parallel manipulator and they do not interfere in the motion of the moving platform. However, they give dexterity to the legs of the parallel manipulator when the base changes its geometry. This type of parallel manipulator can operate in confined environments where stiffness, accuracy and force are necessary and the change of the robot geometry permits dexterity and agility so that the parallel manipulator can adapt to different Abstract This paper presents a new class of parallel manipulators called variable-configuration parallel manipulators with self-aligning. These parallel manipulators can change their form to carry out different tasks in different places, this change being managed by additional degrees-of-freedom of self-aligning. These degrees-of-freedom of self-aligning do not interfere in the motion of the moving platform, they are passive/null degrees-of-freedom that only permit changes in the form of the parallel manipulator. As an example, we present a new 3-DOF translational variable-configuration parallel manipulator with self-aligning. We develop the mobility analysis, the workspace volume analysis, the position analysis and the velocity analysis of this new 3-DOF parallel manipulator.