Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

An approach to quantify manual expertise with collaborative robotics in mind

Nassim Benhabib 1 Vincent Padois 1 David Daney 1
1 AUCTUS - Augmenting human comfort in the factory using cobots
Inria Bordeaux - Sud-Ouest, Bordeaux INP - Institut Polytechnique de Bordeaux
Abstract : This article presents a quantification approach of manual expertise. The long term goal is to better understand the notion of manual expertise in order to improve the design of collaborative robots, both from a mechanical and control point of view. Based on the existing literature and through exchanges with highly skilled manual operators, we first propose a definition of manual expertise. Based on this definition, we propose quantitative evaluation criteria for three important dimensions of a manual task completion: safety, discomfort and performance. These criteria are evaluated in experiments relying on a physically realistic mock-up of a wood milling task, a highly demanding task in the carpentry domain. This mock-up includes a 7-DOF collaborative robot used both to reproduce the cutting tool wrenches and provide measurement of the wooden part motion. This experimental setup is used in a training protocol including two groups of 5 novice subjects. This protocol confirms that the proposed approach allows to observe and analyse the handling strategy developed by an operator through training as well as to correlate the type of training to the nature of the developed expertise.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.inria.fr/hal-03172351
Contributor : Nassim Benhabib <>
Submitted on : Wednesday, March 17, 2021 - 4:16:32 PM
Last modification on : Thursday, March 18, 2021 - 1:16:05 PM

File

IROS_2021_Benhabib_Nassil.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03172351, version 1

Collections

Citation

Nassim Benhabib, Vincent Padois, David Daney. An approach to quantify manual expertise with collaborative robotics in mind. 2021. ⟨hal-03172351⟩

Share

Metrics

Record views

37

Files downloads

159