Kinematic and kinetic performance are important issues in designing multi-degree-of-freedom mechanisms such as robotic manipulators. In the engineering design stage, it is especially important that the designer can grasp the characteristics of the mechanism. The aim of this study is to develop a means of representing the kinematic and kinetic performance of the mechanism in such a way that the performance characteristics are quantified analytically and visible graphically to the designer in their entirety at the conceptual design stage. Various performance indices are derived from the Jacobian matrix and its quadratic form. These performance indices are the local kinematic cross-coupling index (angle of intersection between column vectors of the Jacobian), the local directional mobility index (ratio of Jacobian’s eigenvalues), and the local efficiency index (product of Jacobian’s eigenvalues). Graphical images of these performance characteristics using eigen-ellipsoid and workspace trajectory contours are introduced. Critical performance points in mechanism workspace are identified and elaborated for design considerations. Based on the graphical representation of these performance characteristics, design rules for achieving different performance objectives can easily be implemented. This method is applicable to computer-aided design of a mechanism and predetermination of its kinematic and kinetic performance.

No abstract

Overnight polysomnography (PSG) is currently the standard diagnostic procedure for obstructive sleep apnea (OSA). It has been known that monitoring of head position in sleep is crucial not only for the diagnosis (positional sleep apnea) but also for the management of OSA (positional therapy). However, there are no sensor systems available clinically to hook up with PSG for accurate head position monitoring. In this paper, an accelerometer-based sensing system for accurate head position monitoring is developed and realized. The core CORDIC- (COordinate Rotation DIgital Computer-) based tilting sensing algorithm is realized in the system to quickly and accurately convert accelerometer raw data into the desired head position tilting angles. The system can hook up with PSG devices for diagnosis to have head position information integrated with other PSG-monitored signals. It has been applied in an IRB test in Taipei Veterans General Hospital and has been proved that it can meet the medical needs of accurate head position monitoring for PSG diagnosis.

Analytical and graphical performance synthesis tools for multi-DOF mechanisms are developed in the companion of this two-part paper [1]. Various performance indices derived from the Jacobian matrix for analyzing performance characteristics of multi-DOF mechanisms were proposed. These performance indices are the local kinematic coupling index (inner product of the Jacobian column vectors), the local directional mobility index (ratio of the Jacobian’s eigenvalues), and the local efficiency index (product of the Jacobian’s eigenvalues). In this paper, effort is placed on the applicability of the proposed analytical and graphical synthesis tools used to aid the design of multi-DOF mechanisms. Two examples representing open and closed chain mechanisms will be used to illustrate the effectiveness and efficiency of proposed method. First, a two-link planar robotic manipulator is used to apply the the proposed method. Following that, a two DOF parallel drive road simulator illustrates applicability of the suggested tools. Through these examples, geometric parameters and joint range limits of the mechanism are optimized through the evaluation of performance indices, eigen-ellipse and intersection angle between trajectory contours.