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

-- Existing multi - finger robotic hands face several limitations, including excessive weight, mechanical complexity, high cost, and constraints in both payload capacity and de - grees of freedom (DoF). These challenges hinder their wide adoption, especially when paired with collaborative robotic arms with limited payload capacity. To address these challenges, we present 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. 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. HE rise of automation in recent decades has funda - mentally transformed modern manufacturing, delivering greater efficiency, reduced costs, and increased adaptability [1]. However, due to software complexity, hardware con - straints, and limited adaptability, assembly floors have been the least beneficiaries of automation. The technological lag in assembly automation, partly rooted in Moravec's paradox, stems primarily from the technical complexity of even the simplest tasks, such as threading a wire through a hole or connecting an electrical plug [2], let alone assembling intricate parts. On the software front, with recent advances in machine learning, the assimi - lation of large volumes of multi - modal sensory data and the generation of high - dimensional actions is now more feasible than ever before [5], [6].