Investigating the Fitness of Finger Grippers for Dynamic Tactile Manipulation Under Static Object Conditions

Abstract

Robotic system development must adopt a holistic approach for tactile and dynamic tasks, shifting from the decoupled design of end-effectors and robot manipulators for traditional sequential tasks. Although established metrics exist for traditional tasks, such as pick-and-place, they lack the nuanced evaluation required for dynamic and tactile operations. Accordingly, this paper introduces an integrated framework that defines and unifies decoupled and coupled gripper metrics into a single perspective. We categorise gripper metrics based on their interaction with the robot manipulator, which can be entirely decoupled, coupled by time-sequence, or coupled. Using this classification, we propose 16 metrics to evaluate force control, force reaction, and efficiency. We introduce three new experimental setups and describe the corresponding procedures to quantify these metrics. Results from three commercial finger grippers demonstrate the efficacy of the proposed metrics, revealing each gripper’s strengths and limitations when integrated into different manipulator systems. Incorporating these metrics into performance reviews provides a comprehensive evaluation of robotic system fitness, considering dynamic, real-time challenges. This supports informed design choices and enhances tactile manipulation tasks.