Dimensional synthesis of kinematically redundant serial manipulators for cluttered environments

A systematic procedure for synthesizing kinematically redundant serial manipulators is proposed in this paper. For a given cluttered workcell, the task space locations (TSL’s) for the desired manipulator are prescribed. The synthesis is performed with the objective of reachability of the manipulator at specified TSL’s, while avoiding obstacles. The problem is formulated as a constrained optimization problem, minimizing the positional error and simultaneously avoiding any collision of the manipulator with either the obstacles or within its links. The technique used to solve the resulting constrained optimization problem is the classical Augmented Lagrangian Method. The paper presents a discussion on the past works in this field. It is observed that the presented literature is confined to special cases only while the proposed method involves full generality of the synthesis problem. The availability of such an algorithm working for full generality is important, particularly for highly constrained environments. The efficiency of the proposed approach to synthesize the desired redundant manipulators is exhibited through diverse cases. The resulting synthesized manipulators are further checked for the possibility of feasible paths between TSL’s. An outline of the development of a redundant manipulator, synthesized through the presented approach, is also included in this paper.