Krysalis Hand: A Lightweight, High-Payload, 18-DoF Anthropomorphic End-Effector for Robotic Learning and Dexterous Manipulation

Al Arsh Basheer, Justin S. Chang, Yuyang Chen, David Kim, Iman Soltani

Abstract

This paper presents the Krysalis Hand, a five-finger robotic end-effector that combines a lightweight design, high payload capacity, and a high number of degrees of freedom (DoF) to enable dexterous manipulation in both industrial and research settings. This design integrates the actuators within the hand while maintaining an anthropomorphic form. Each finger joint features a self-locking mechanism that allows the hand to sustain large external forces without active motor engagement. This approach shifts the payload limitation from the motor strength to the mechanical strength of the hand, allowing the use of smaller, more cost-effective motors. With 18 DoF and weighing only 790 grams, the Krysalis Hand delivers an active squeezing force of 10 N per finger and supports a passive payload capacity exceeding 10 lbs. These characteristics make Krysalis Hand one of the lightest, strongest, and most dexterous robotic end-effectors of its kind. Experimental evaluations validate its ability to perform intricate manipulation tasks and handle heavy payloads, underscoring its potential for industrial applications as well as academic research.

Introduction

In robotic manipulation, researchers often face a "Goldilocks" problem regarding end-effectors. Simple industrial grippers are robust but lack the dexterity required for complex assembly. Conversely, high-DoF anthropomorphic hands are capable but often prohibitively expensive, heavy (sometimes exceeding 4kg), and mechanically fragile. This weight issue is critical for collaborative robots (cobots) with limited payload capacities. The Krysalis Hand addresses these challenges by providing a high-DoF, lightweight, and self-locking design compatible with low-cost robotic arms.

Methodology and Design

The Krysalis Hand employs several engineering strategies to balance weight and performance:

  • Leadscrew Actuation: Instead of maintenance-heavy tendon systems, the hand uses motor-driven leadscrews integrated directly into the fingers.
  • Self-Locking Mechanism: The leadscrew geometry ensures that external forces cannot "back-drive" the motor, allowing the hand to hold heavy objects indefinitely without drawing power.
  • Coupled Abduction-Adduction: To optimize palm space, the side-to-side movement of the index, ring, and pinky fingers is coupled to a single servo motor, reflecting biological constraints where these joints rarely move independently during functional grasps.
  • Hybrid Fabrication: The hand is primarily 3D printed using stereolithography (SLA) with "Tough 1500" resin, while high-stress components like the thumb's CMC joint are metal 3D printed using an aluminum alloy.
The Krysalis Hand mounted on a robotic arm, demonstrating its compact 18-DoF anthropomorphic design.

To ensure human-like performance, the design was benchmarked against the US Army Hand Anthropometry dataset for proportions and human Range of Motion (RoM) data for joint flexibility.

Key Insights

  • Mechanical Strength over Motor Torque: By utilizing self-locking leadscrews, the payload limit is determined by material strength rather than motor torque. This allows for smaller, more cost-effective motors without sacrificing lifting power.
  • Anthropomorphism for Learning: The 18-DoF design aligns the robot's action space with a human's, simplifying "Learning from Demonstration" (LfD) as human movements can be mapped 1:1 to the robot via teleoperation.

Experimental Results

The Krysalis Hand demonstrated significant performance across several metrics:

  • Weight & Payload: The entire assembly weighs only 790 grams but can passively hold a 10 lb (4.54 kg) dumbbell.
  • Active Force: Each finger exerts an active squeezing force of 10 N.
  • Dexterity: The hand achieved an average 11.2% greater range of motion than the average human hand across its joints.
  • Teleoperation: Real-time teleoperation using motion-capture gloves successfully demonstrated various grasp types, from heavy power tools to delicate precision tasks.
The Krysalis Hand's versatility demonstrated through high-payload tool use, human-like gestures, and real-time teleoperation.

Conclusion and Future Work

The Krysalis Hand successfully bridges the gap between high-end research end-effectors and affordable industrial grippers. While currently limited by resin-based material strength and wrist extension range, future iterations will explore fully metal constructions and integrated tactile sensing to further enhance autonomous manipulation capabilities.

Summary updated

Abstract

This paper presents the Krysalis Hand, a five-finger robotic end-effector that combines a lightweight design, high payload capacity, and a high number of degrees of freedom (DoF) to enable dexterous manipulation in both industrial and research settings. This design integrates the actuators within the hand while maintaining an anthropomorphic form. Each finger joint features a self-locking mechanism that allows the hand to sustain large external forces without active motor engagement. This approach shifts the payload limitation from the motor strength to the mechanical strength of the hand, allowing the use of smaller, more cost-effective motors. With 18 DoF and weighing only 790 grams, the Krysalis Hand delivers an active squeezing force of 10 N per finger and supports a passive payload capacity exceeding 10 lbs. These characteristics make Krysalis Hand one of the lightest, strongest, and most dexterous robotic end-effectors of its kind. Experimental evaluations validate its ability to perform intricate manipulation tasks and handle heavy payloads, underscoring its potential for industrial applications as well as academic research. All code related to the Krysalis Hand, including control and teleoperation, is available on the project GitHub repository: https://github.com/Soltanilara/Krysalis_Hand
BibTeX
@misc{basheer2025krysalis,
  author = {Al Arsh Basheer and Justin S. Chang and Yuyang Chen and David Kim and Iman Soltani},
  title = {{Krysalis Hand: A Lightweight, High-Payload, 18-DoF Anthropomorphic End-Effector for Robotic Learning and Dexterous Manipulation}},
  year = {2025},
  eprint = {2504.12967},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2504.12967},
}
Figure