Design Optimization of Serial Link Redundant Manipulator: An Approach Using Global Performance Metric

In this paper, an approach for design optimization of serial-link redundant manipulator is presented with exemplar optimizations on a 3-R planar and a 5-R spatial manipulator. The optimization is based on a global kinematic performance metric, which is chosen to be the Global Conditioning Index (GCI) based on condition number of manipulator Jacobian. The workspaces of 3-R planar and 5-R spatial manipulators are numerically computed using Monte Carlo technique in task space, which is utilized in evaluating the GCI. An algorithm is developed to compute the inverse kinematics of the 3-R planar and the 5-R spatial manipulators in finding the workspace and in turn the optimized solutions. Finally, design optimization problems are formulated for the 3-R and 5-R manipulator cases using the GCI as the objective function and solved (orientation of the end-effector is not included here). Further, genetic algorithm based method is used to solve the problem in validating the optimization results.