Modeling and Learning Constraints for Creative Tool Use

Improvisation is a hallmark of human creativity and serves a functional purpose in completing everyday tasks with novel resources. This is particularly exhibited in tool-using tasks: When the expected tool for a task is unavailable, humans often are able to replace the expected tool with an atypical...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Tesca Fitzgerald , Ashok Goel , Andrea Thomaz
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	tool manipulation tool transfer learning from corrections human-robot interaction cognitive robotics Mechanical engineering and machinery TJ1-1570 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/915cd4e2771a44da8b12a761ae5d5013
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:915cd4e2771a44da8b12a761ae5d5013
record_format	dspace
spelling	oai:doaj.org-article:915cd4e2771a44da8b12a761ae5d50132021-11-05T06:25:56ZModeling and Learning Constraints for Creative Tool Use2296-914410.3389/frobt.2021.674292https://doaj.org/article/915cd4e2771a44da8b12a761ae5d50132021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/frobt.2021.674292/fullhttps://doaj.org/toc/2296-9144Improvisation is a hallmark of human creativity and serves a functional purpose in completing everyday tasks with novel resources. This is particularly exhibited in tool-using tasks: When the expected tool for a task is unavailable, humans often are able to replace the expected tool with an atypical one. As robots become more commonplace in human society, we will also expect them to become more skilled at using tools in order to accommodate unexpected variations of tool-using tasks. In order for robots to creatively adapt their use of tools to task variations in a manner similar to humans, they must identify tools that fulfill a set of task constraints that are essential to completing the task successfully yet are initially unknown to the robot. In this paper, we present a high-level process for tool improvisation (tool identification, evaluation, and adaptation), highlight the importance of tooltips in considering tool-task pairings, and describe a method of learning by correction in which the robot learns the constraints from feedback from a human teacher. We demonstrate the efficacy of the learning by correction method for both within-task and across-task transfer on a physical robot.Tesca Fitzgerald Ashok Goel Andrea Thomaz Frontiers Media S.A.articletool manipulationtool transferlearning from correctionshuman-robot interactioncognitive roboticsMechanical engineering and machineryTJ1-1570Electronic computers. Computer scienceQA75.5-76.95ENFrontiers in Robotics and AI, Vol 8 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	tool manipulation tool transfer learning from corrections human-robot interaction cognitive robotics Mechanical engineering and machinery TJ1-1570 Electronic computers. Computer science QA75.5-76.95
spellingShingle	tool manipulation tool transfer learning from corrections human-robot interaction cognitive robotics Mechanical engineering and machinery TJ1-1570 Electronic computers. Computer science QA75.5-76.95 Tesca Fitzgerald Ashok Goel Andrea Thomaz Modeling and Learning Constraints for Creative Tool Use
description	Improvisation is a hallmark of human creativity and serves a functional purpose in completing everyday tasks with novel resources. This is particularly exhibited in tool-using tasks: When the expected tool for a task is unavailable, humans often are able to replace the expected tool with an atypical one. As robots become more commonplace in human society, we will also expect them to become more skilled at using tools in order to accommodate unexpected variations of tool-using tasks. In order for robots to creatively adapt their use of tools to task variations in a manner similar to humans, they must identify tools that fulfill a set of task constraints that are essential to completing the task successfully yet are initially unknown to the robot. In this paper, we present a high-level process for tool improvisation (tool identification, evaluation, and adaptation), highlight the importance of tooltips in considering tool-task pairings, and describe a method of learning by correction in which the robot learns the constraints from feedback from a human teacher. We demonstrate the efficacy of the learning by correction method for both within-task and across-task transfer on a physical robot.
format	article
author	Tesca Fitzgerald Ashok Goel Andrea Thomaz
author_facet	Tesca Fitzgerald Ashok Goel Andrea Thomaz
author_sort	Tesca Fitzgerald
title	Modeling and Learning Constraints for Creative Tool Use
title_short	Modeling and Learning Constraints for Creative Tool Use
title_full	Modeling and Learning Constraints for Creative Tool Use
title_fullStr	Modeling and Learning Constraints for Creative Tool Use
title_full_unstemmed	Modeling and Learning Constraints for Creative Tool Use
title_sort	modeling and learning constraints for creative tool use
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/915cd4e2771a44da8b12a761ae5d5013
work_keys_str_mv	AT tescafitzgerald modelingandlearningconstraintsforcreativetooluse AT ashokgoel modelingandlearningconstraintsforcreativetooluse AT andreathomaz modelingandlearningconstraintsforcreativetooluse
_version_	1718444483701374976

Modeling and Learning Constraints for Creative Tool Use

Ejemplares similares