No abstract

Mechanism synthesis, the identification of the parameters of a mechanism, has been extensively studied especially for four-bar linkages using graphical and numerical optimization approaches. Graphical techniques follow a number of predefined steps and rely heavily on the user. Numerical optimization techniques that require the user to provide ''good initial guesses'' or bounds for the design variables have also been applied. In general, a linkage is synthesized for function generation, motion generation, and path generation. This article studies four-bar mechanism synthesis by combining Differential Evolution, an evolutionary optimization scheme that can search outside the initial defined bounds for the design variables, and a newly developed novel technique called the Geometric Centroid of Precision Points (GCPP) and the distant precision point in defining the initial bounds for the design variables. The developed methodology has been applied to the synthesis of four-bar linkages for path generation with prescribed timing, where the coupler point is required to pass through a number of precision points within a prescribed accuracy level and in the correct order, and for the generation of families of coupler curves. Two penalty functions were used, one for constraint violation and one for relative accuracy. The results of the application of this approach could also be used as ''good initial guesses'' for improving the desired accuracy level. Examples demonstrating the successful application of the developed methodology are presented.

In this paper, we discuss optimum robot design based on task specifications using evolutionary optimization approaches. The three evolutionary optimization approaches employed are Simple Genetic Algorithms, Genetic Algorithms with elitism, and Differential Evolution. These approaches were used for the optimum design of SCARA and articulated type manipulators. The objective function minimizes the torque required for the motion subject to deflection and physical constraints with the design variables being the physical characteristics of link (length and cross sectional area parameters). In this work, we experimented links with various cross sections. The main findings of this research are that the differential evolution converges quickly, requires significantly less number of iterations and achieves better results.

In Part I of this paper, we present the development of tools for engineering design, analysis and simulation over the World Wide Web accessible via a web browser. The tools utilize the power and versatility provided by the software package MATLAB and the features of the web toolbox. This work aims at improving the understanding of engineering fundamentals through interactive real-time simulation. The current environment consists of a number of modules for the interactive real-time simulation of linear time invariant dynamic systems and for analysis and synthesis of planar mechanisms. The properly designed input interface facilitates the easy understanding of the required inputs from the user. The returned information to the user's web browser consists of ASCII (textual), graphical, and pictorial representations, and provides information of the input and solutions. In addition, where appropriate, the user has the option to view an animation of the system. This work demonstrates advantages of using this technology to aid in classroom instruction, ease of implementation for application in other domains, and provides valuable analysis tools to students and engineering professionals. Examples will be presented in Part II to demonstrate the developed tools. Users with access to the Internet can use the developed mechanism synthesis and analysis modules at http://zodhia.uta.edu/development. ß

In this paper, we discuss optimum robot design based on task specifications using evolutionary optimization approaches. The three evolutionary optimization approaches employed Simple Genetic Algorithms, Genetic Algorithms with elitism, and Differential Evolution. These approaches were used for the optimum design of SCARA and articulated type manipulators. The objective function minimizes the torque required for the motion subject to deflection and physical constraints with the design variables being the link physical characteristics (length and cross sectional area parameters). In this work, we experimented with link various cross sections. The main findings of this research are that the differential evolution converges quickly, requires significantly less number of iterations and achieves better results.