DeGrip: A Compact Cable-driven Robotic Gripper for Desktop Disassembly

TL;DR

The paper tackles the challenge of automating end-of-life (EOL) desktop disassembly by introducing DeGrip, a $3$-DOF cable-driven gripper designed for confined spaces. It decouples wrist and jaw actuation to enable arbitrary orientations while maintaining a compact form factor, with a drive system that keeps actuators away from the gripper tip. A physics-based disassembly platform in Isaac Sim is developed to validate DeGrip across diverse desktop configurations, including RAM, SSD, and HDD components in tight or arbitrary poses. The work demonstrates DeGrip’s potential to enable learning-based disassembly policies and data collection for automated recycling, with future work focusing on hardware enhancements, real-world testing, and sim-to-real policy transfer.

Abstract

Intelligent robotic disassembly of end-of-life (EOL) products has been a long-standing challenge in robotics. While machine learning techniques have shown promise, the lack of specialized hardware limits their application in real-world scenarios. We introduce DeGrip, a customized gripper designed for the disassembly of EOL computer desktops. DeGrip provides three degrees of freedom (DOF), enabling arbitrary configurations within the disassembly environment when mounted on a robotic manipulator. It employs a cable-driven transmission mechanism that reduces its overall size and enables operation in confined spaces. The wrist is designed to decouple the actuation of wrist and jaw joints. We also developed an EOL desktop disassembly environment in Isaac Sim to evaluate the effectiveness of DeGrip. The tasks were designed to demonstrate its ability to operate in confined spaces and disassemble components in arbitrary configurations. The evaluation results confirm the capability of DeGrip for EOL desktop disassembly.

Figures (10)

• Figure 1: DeGrip: A customized gripper for the disassembly of EOL desktops. We propose a customized gripper design and create simulation environments replicating EOL desktops. We then validate the effectiveness of the customized gripper within the simulation environment.
• Figure 2: Overview of the DeGrip design, with physical prototype. A. Gripper links: base, wrist, gripper 1, and gripper 2. B. Yaw and pitch DOFs of the gripper. C. The servo motors drive the joints, and the mounting plate attaches the gripper to robotic manipulators. D-F. Range of motion of gripper joints: D. wrist, E. jaw 1, and F. jaw 2. G-I. Photos of the gripper prototype.
• Figure 3: Cable transmission design of DeGrip, with physical prototype. A. Cable-driven transmission mechanism. The cables are driven by the drive modules and pass through the cable guides. Cables for the jaws additionally pass through guide caps to establish a decoupled movement between joints. B-C. The length of the jaw cables remains the same as the wrist rotates, thanks to the guide caps on the wrist joint. D-F. Photos of the gripper prototype. D. gripper with covers removed, showing the cables. E. Close-up photo of the jaws. F. Cable length change is negligible as the wrist rotates.
• Figure 4: Tensioning process for the cable-driven transmission module of the DeGrip. The capstan upper piece is first turned to pre-tension the cable. Then the pieces are secured with screws and locked in place by ratchets. This completes the drive module.
• Figure 5: DeGrip mounted on the robot, in reality and in simulation. A. Reality. B. Simulation.