Enhanced Dexterity Maps (EDM): A New Map for Manipulator Capability Analysis

The ability of a manipulator to compute a geometry-aware quality index for general tasks with different joint configurations is essential. Such workspace assessment is a well-known and studied field in existing robotics literature, often deployed through embodied structures such as voxelized maps. Notwithstanding, existing literature solely focuses on the assessment of a single pose (end-effector), neglecting the whole-body structure and its dexterity, which allows for secondary task optimization, nullspace motion, body placement, and improved manipulability. The proposed enhanced dexterity maps (EDM) aims to close these gaps using an augmented data structure. It offers a systematic analysis of disjoint flip solutions and accommodates additional performance metrics. Further analysis of EDMs through case studies highlight the limitations of existing methods and supports the need for a whole-body analysis.