Embodied Manipulation with Past and Future Morphologies through an Open Parametric Hand Design

TL;DR

An open parametric design that integrates techniques for simplified customization, fabrication, and control with design features to maximize behavioral diversity is introduced that enables rapid production of single-piece 3D-printable hands without compromising dexterous behaviors.

Abstract

A human-shaped robotic hand offers unparalleled versatility and fine motor skills, enabling it to perform a broad spectrum of tasks with precision, power and robustness. Across the paleontological record and animal kingdom we see a wide range of alternative hand and actuation designs. Understanding the morphological design space and the resulting emergent behaviors can not only aid our understanding of dexterous manipulation and its evolution, but also assist design optimization, achieving, and eventually surpassing human capabilities. Exploration of hand embodiment has to date been limited by inaccessibility of customizable hands in the real-world, and by the reality gap in simulation of complex interactions. We introduce an open parametric design which integrates techniques for simplified customization, fabrication, and control with design features to maximize behavioral diversity. Non-linear rolling joints, anatomical tendon routing, and a low degree-of-freedom, modulating, actuation system, enable rapid production of single-piece 3D printable hands without compromising dexterous behaviors. To demonstrate this, we evaluated the design's low-level behavior range and stability, showing variable stiffness over two orders of magnitude. Additionally, we fabricated three hand designs: human, mirrored human with two thumbs, and aye-aye hands. Manipulation tests evaluate the variation in each hand's proficiency at handling diverse objects, and demonstrate emergent behaviors unique to each design. Overall, we shed light on new possible designs for robotic hands, provide a design space to compare and contrast different hand morphologies and structures, and share a practical and open-source design for exploring embodied manipulation.

Figures (11)

• Figure 1: The open parametric design for accessible, behaviorally diverse, functional hands. (A) Utilising single-piece 3D printing with parametric design, anatomical tendon routing/joints, and low degree of freedom synergistic actuation hands can be rapidly customized to suit a variety of tasks. (B) Hand behavioral diversity expands task capabilities and emerges from complex interactions between the non-linear joints, tendon transmissions/joint coupling effects, and modulation of these by actuators. (C) Functionality from accessible design/manufacturing, practical uses with high strength/capabilities in a variety of tasks, and minimal actuation.
• Figure 2: Morphological design space enabled by open parametric design. (A) Exploration of embodied manipulation with short term developmental axis and long-term evolutionary axis (Primate hands from almecija2017hands, early hominid Ardipithecus ramidus 4.4 million years ago white2009ardipithecus). Distance from centre (full-scale modern human hand) measures dissimilarity. Study of 'natural' hand variations can aid development of more capable 'artificial' designs. (B) Highlighted human, aye-aye and two thumbed/mirrored hands are fabricated and prove range of design.
• Figure 3: Parametric design for emerging complexity with stable behaviors and single-piece 3D printing capabilities. (A) Single ligament, non-linear, rolling contact joints assembled into bones, then fingers, then hands. (B) Highly customizable parameters from low-level joint parameters for stiffness/strength/range to high-level finger shapes and distributions for workspace and dexterity. (C) Joint design constraints. To prevent joint slipping, net tensile force should be sustained in ligament. With relative pulley placements $l_1 >> l_2$, the joint is less likely to slip (stable). (D) Additive manufacturing defects minimized in printing plane, thereby constraining relative finger alignment and print angle. (E) 3D printing example: single-piece print process with support material, tendon routing, then tendon pre-tensioning for regular joint operation.
• Figure 4: Joint and finger range of passive behaviors with varying tendon configuration. (A) joint angle and finger workspace compared to human data. (B) Joint/finger stiffness range with varying test axes (three sub figures), start pose, and spring stiffness (solid and dotted lines). Single variable stiffness change up to 290%, multi-variable up to 660%. Finger starting poses controlled by varying relative tendon lengths.
• Figure 5: Joint and finger stability and range of behavior with varying parametric design. (A) Stiffness range in maximum and minimum stiffness axes (Fig. \ref{['fig:4']}D) with varied finger design. Up to 400% change in stiffness by varying finger design. (B) Additional tendons can be added to increase thumb opposition force/stiffness.