A method has been developed that successfully represents the workspace of a parallel manipulator within a finite twist image space. A point in this space represents a unique position and orientation of the end effector. The method of analysis is based upon the established technique of simplifying the parallel manipulator, by modelling each leg as an independent serial manipulator. The workspace corresponding to each serial manipulator is mapped onto the image space to produce a hyper-volume. The intersection of the individual hyper-volumes represents the workspace of the complete parallel manipulator. Since the hyper-volume corresponds to all possible positions attainable by the end effector, this represents the reachable workspace. Within the reachable workspace there lies subsets of volumes in ⮲3 that correspond to all possible orientations attainable. Such volumes represent the dextrous workspace. Although the method is illustrated by the use of a Stewart platform, it is equally applicable to the general parallel manipulator. The method is demonstrated successfully by the use of a 3 legged, 3-DOF planar parallel manipulator.

This content is only available via PDF.
You do not currently have access to this content.