Grasping the Performance: Facilitating Replicable Performance Measures via Benchmarking and Standardized Methodologies

Published

December 31, 2015

Author(s)

Joseph A. Falco, Karl Van Wyk, Shuo Liu, Stefano Carpin

Abstract

The expectation for robots to work side-by-side with humans in unstructured environments while reliably grasping and manipulating objects is a distant reality. Multiple reasons account for this persistent gap - some are technological, whereas others are methodological. We postulate that a major obstacle inhibiting progress in this area surrounds the continued inability to replicate and compare results generated by the grasping community. With a surge of robotic hand designs and algorithmic developments, it is imperative to be able to make informed decisions and draw quantitative conclusions regarding the performance of robotic hands in various task-related settings. Standardized performance testing is an emerging tool within the robotics community that provides evaluations to help match capabilities to end-user needs while also providing developers insight for improving their product designs. This manuscript introduces test methods for evaluation of hardware and control benchmarks that are demonstrated using a set of laboratory robotic hand platforms. Resulting measures are presented that quantitatively describe hand performance enabling the analysis of their capabilities.

Citation

IEEE Robotics and Automation Society Robotics and Automation Magazine - Special Issue on Replicable and Measurable Robotics Research

Pub Type

Others

Download Paper

Local Download

Keywords

Robotic Hands, Grasping, Performance Measures and Benchmarking, Force and Tactile Sensing, Force Control

Citation

Falco, J. , Van, K. , Liu, S. and Carpin, S. (2015), Grasping the Performance: Facilitating Replicable Performance Measures via Benchmarking and Standardized Methodologies, IEEE Robotics and Automation Society Robotics and Automation Magazine - Special Issue on Replicable and Measurable Robotics Research, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=917697 (Accessed December 8, 2023)