2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction

Veronica E. Arriola-Rios; Jeremy Wyatt

doi:10.1007/978-1-4471-2318-7_14

2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction

Veronica E. Arriola-Rios^*, Jeremy Wyatt

^*Corresponding author for this work

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Deformable objects abound in nature, and future robots must be able to predict how they are going to behave in order to control them. In this paper we present a method capable of learning to predict the behaviour of deformable objects. We use a mass-spring-like model, which we extended to better suit our purposes, and apply it to the concrete scenario of robotic manipulation of an elastic deformable object. We describe a procedure for automatically calibrating the parameters for the model taking images and forces from a real sponge as ground truth. We use this ground truth to provide error measures that drive an evolutionary process that searches the parameter space of the model. The resulting calibrated model can make good predictions for 200 frames (6.667 seconds of real time video) even when tested with forces being applied in different positions to those trained.

Original language	English
Title of host publication	Research and Development in Intelligent Systems XXVIII
Subtitle of host publication	Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence
Editors	Max Bramer, Miltos Petridis, Lars Nolle
Publisher	Springer
Pages	195-208
Number of pages	14
Edition	1
ISBN (Electronic)	9781447123187
ISBN (Print)	9781447123170
DOIs	https://doi.org/10.1007/978-1-4471-2318-7_14
Publication status	Published - 17 Nov 2011
Event	31st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011 - Cambridge, United Kingdom Duration: 13 Dec 2011 → 15 Dec 2011

Conference

Conference	31st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011
Country/Territory	United Kingdom
City	Cambridge
Period	13/12/11 → 15/12/11

Keywords

Ground Truth
Force Sensor
Finite Element Method Model
Graphical Constraint
Deformable Object

ASJC Scopus subject areas

Artificial Intelligence
Information Systems

Access to Document

10.1007/978-1-4471-2318-7_14

FP7_Collab_CogX: Cognitive Systems that Self-Understand and Self-Extend
Wyatt, J., Dearden, R. & Sloman, A.
European Commission, European Commission - Management Costs
1/05/08 → 30/06/12
Project: Research

Cite this

Arriola-Rios, V. E., & Wyatt, J. (2011). 2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction. In M. Bramer, M. Petridis, & L. Nolle (Eds.), Research and Development in Intelligent Systems XXVIII: Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence (1 ed., pp. 195-208). Springer. Advance online publication. https://doi.org/10.1007/978-1-4471-2318-7_14

Arriola-Rios, Veronica E. ; Wyatt, Jeremy. / 2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction. Research and Development in Intelligent Systems XXVIII: Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence. editor / Max Bramer ; Miltos Petridis ; Lars Nolle. 1. ed. Springer, 2011. pp. 195-208

@inproceedings{230c451366a84189a4a8d49653f79ceb,

title = "2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction",

abstract = "Deformable objects abound in nature, and future robots must be able to predict how they are going to behave in order to control them. In this paper we present a method capable of learning to predict the behaviour of deformable objects. We use a mass-spring-like model, which we extended to better suit our purposes, and apply it to the concrete scenario of robotic manipulation of an elastic deformable object. We describe a procedure for automatically calibrating the parameters for the model taking images and forces from a real sponge as ground truth. We use this ground truth to provide error measures that drive an evolutionary process that searches the parameter space of the model. The resulting calibrated model can make good predictions for 200 frames (6.667 seconds of real time video) even when tested with forces being applied in different positions to those trained.",

keywords = "Ground Truth, Force Sensor, Finite Element Method Model, Graphical Constraint, Deformable Object",

author = "Arriola-Rios, {Veronica E.} and Jeremy Wyatt",

year = "2011",

month = nov,

day = "17",

doi = "10.1007/978-1-4471-2318-7_14",

language = "English",

isbn = "9781447123170",

pages = "195--208",

editor = "Max Bramer and Miltos Petridis and Lars Nolle",

booktitle = "Research and Development in Intelligent Systems XXVIII",

publisher = "Springer",

edition = "1",

note = "31st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011 ; Conference date: 13-12-2011 Through 15-12-2011",

}

Arriola-Rios, VE & Wyatt, J 2011, 2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction. in M Bramer, M Petridis & L Nolle (eds), Research and Development in Intelligent Systems XXVIII: Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence. 1 edn, Springer, pp. 195-208, 31st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011, Cambridge, United Kingdom, 13/12/11. https://doi.org/10.1007/978-1-4471-2318-7_14

2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction. / Arriola-Rios, Veronica E.; Wyatt, Jeremy.
Research and Development in Intelligent Systems XXVIII: Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence. ed. / Max Bramer; Miltos Petridis; Lars Nolle. 1. ed. Springer, 2011. p. 195-208.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - 2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction

AU - Arriola-Rios, Veronica E.

AU - Wyatt, Jeremy

PY - 2011/11/17

Y1 - 2011/11/17

N2 - Deformable objects abound in nature, and future robots must be able to predict how they are going to behave in order to control them. In this paper we present a method capable of learning to predict the behaviour of deformable objects. We use a mass-spring-like model, which we extended to better suit our purposes, and apply it to the concrete scenario of robotic manipulation of an elastic deformable object. We describe a procedure for automatically calibrating the parameters for the model taking images and forces from a real sponge as ground truth. We use this ground truth to provide error measures that drive an evolutionary process that searches the parameter space of the model. The resulting calibrated model can make good predictions for 200 frames (6.667 seconds of real time video) even when tested with forces being applied in different positions to those trained.

AB - Deformable objects abound in nature, and future robots must be able to predict how they are going to behave in order to control them. In this paper we present a method capable of learning to predict the behaviour of deformable objects. We use a mass-spring-like model, which we extended to better suit our purposes, and apply it to the concrete scenario of robotic manipulation of an elastic deformable object. We describe a procedure for automatically calibrating the parameters for the model taking images and forces from a real sponge as ground truth. We use this ground truth to provide error measures that drive an evolutionary process that searches the parameter space of the model. The resulting calibrated model can make good predictions for 200 frames (6.667 seconds of real time video) even when tested with forces being applied in different positions to those trained.

KW - Ground Truth

KW - Force Sensor

KW - Finite Element Method Model

KW - Graphical Constraint

KW - Deformable Object

UR - http://www.scopus.com/inward/record.url?scp=84881533620&partnerID=8YFLogxK

U2 - 10.1007/978-1-4471-2318-7_14

DO - 10.1007/978-1-4471-2318-7_14

M3 - Conference contribution

SN - 9781447123170

SP - 195

EP - 208

BT - Research and Development in Intelligent Systems XXVIII

A2 - Bramer, Max

A2 - Petridis, Miltos

A2 - Nolle, Lars

PB - Springer

T2 - 31st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011

Y2 - 13 December 2011 through 15 December 2011

ER -

Arriola-Rios VE, Wyatt J. 2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction. In Bramer M, Petridis M, Nolle L, editors, Research and Development in Intelligent Systems XXVIII: Incorporating Applications and Innovations in Intelligent Systems XIX Proceedings of AI-2011, the Thirty-first SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence. 1 ed. Springer. 2011. p. 195-208 Epub 2011 Nov 13. doi: 10.1007/978-1-4471-2318-7_14

2D Mass-spring-like model for prediction of a Sponge's behaviour upon robotic interaction

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Projects

FP7_Collab_CogX: Cognitive Systems that Self-Understand and Self-Extend

Cite this