A Bilingual Cognitive Robot that Learns like a Toddler

Ioanna Giorgi; Angelo Cangelosi; Giovanni L. Masala

A Bilingual Cognitive Robot that Learns like a Toddler

Ioanna Giorgi, Angelo Cangelosi, Giovanni L. Masala

Manchester Metropolitan University (MMU)

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

We propose an embodied architecture featuring a developmental agent and a social robot for human-robot verbal engagement at preschool level. Initially, we trained the agent on bilingual acquisition and demonstrated its skill to appropriately detect the spoken content and automatically match the human user’s language. Multilingual robot agents able to understand and communicate contexts in multiple tongues without loss of meaning in translations can greatly affect human cognitive and interaction capabilities. Furthermore, we teach the robot to gradually learn from experience new knowledge not specified at design time and autonomously use it for task-solving without being trained anew. The demonstrated system shows promising potential to becoming a low-cost social companion that can be easily integrated in the human world.

Original language	English
Title of host publication	IEEE Symposium Series on Computational Intelligence (SSCI 2020)
Publisher	IEEE
Publication status	Accepted/In press - 18 Sept 2020

Keywords

brain-inspired architecture
developmental agent
cognitive system
embodiment
bilingual robot
multi-language development
humanlike learning

RAI: Centre for Robotics and Artificial Intelligence
Cangelosi, A. (PI), Lennox, B. (PI), Weightman, A. (CoI), Dennis, L. (CoI), Dixon, C. (CoI), Fisher, M. (CoI), Herrmann, G. (CoI), Dickson, A. (CoI), Lanzon, A. (CoI), Stancu, A. (CoI), Saiani, A. (CoI), Freitas, A. (CoI), Nini, A. (CoI), Voronkov, A. (CoI), Brass, A. (CoI), Vijayaraghavan, A. (CoI), Parslew, B. (CoI), crowther, W. (CoI), Grieve, B. (CoI), Adorno, B. (CoI), Jay, C. (CoI), Wang, C. C. L. (CoI), Todd, C. (CoI), Soutis, C. (CoI), Arsene, C. (CoI), Dresner, D. (CoI), Barrett, E. (CoI), Gowen, E. (CoI), Arvin, F. (CoI), Podd, F. (CoI), Brown, G. (CoI), Reger, G. (CoI), Cooper, G. (CoI), Mairs, H. (CoI), Yin, H. (CoI), Kinloch, I. (CoI), Eleftheriou, I. (CoI), Li, J. (CoI), Carrasco Gomez, J. (CoI), Ainsworth, J. (CoI), Sinha, J. (CoI), Ozanyan, K. (CoI), Smith, K. (CoI), Twomey, K. (CoI), Margetts, L. (CoI), Ren, L. (CoI), Zhang, L. (CoI), Cordeiro, L. (CoI), Rattray, M. (PI), Bissett, M. (CoI), Elliot, M. (CoI), Alvarez, M. (CoI), Luján, M. (CoI), Nabawy BSc, MSc, PhD, MRAeS, SMAIAA, FHEA, M. (CoI), Peek, N. (CoI), Marjanovic, O. (CoI), Dorn, O. (CoI), Dudek, P. (CoI), Green, P. (CoI), Connolly, P. (CoI), Da Silva Bartolo, P. J. (CoI), Gardner, P. (CoI), Martin, P. (CoI), Potluri, P. (CoI), Curtis, R. (CoI), Schmidt, R. (CoI), Banach, R. (CoI), Batista-Navarro, R. T. (CoI), Kaski, S. (CoI), Midson, S. (CoI), Watson, S. (CoI), Holm, S. (CoI), Furber, S. (CoI), Schlegel, V. (CoI), Mirihanage, W. (CoI), Mansell, W. (CoI), Pan, W. (CoI), Sampson, W. (CoI), Sellers, W. (CoI), Yang, W. (CoI), Cai, P. (CoI), Sun, Y. (CoI), Alharthi, A. (Researcher), Macario Rojas, A. (Researcher), Serhan, B. (Researcher), Yu, C. (Researcher), Abara, D. (Researcher), Lopez Pulgarin, E. (Researcher), Faruq, F. (Researcher), Tavella, F. (Researcher), Semeraro, F. (Researcher), Liu, G. (Researcher), Fang, G. (Researcher), Niu, H. (Researcher), Taylor, H. (PI), Zhu, H. (PGR student), Collenette, J. (Researcher), Amano, K. (Researcher), Lo, K. C. J. (PGR student), Raggioli, L. (Researcher), Romeo, M. (Researcher), Ruocco, M. (PGR student), Ghaffari Saadat, M. (Researcher), Walmsley, M. (Researcher), Mubarik, A. (Researcher), Vinanzi, S. (Researcher), Su, Y.-H. (PGR student), McAleese, H. (PGR student), Stringer, P. (PGR student), Stoican, R. (PGR student), Ye, R. (PGR student), Kurawa, S. S. (PGR student), Zhang, T. (PGR student), Krywonos, W. (PGR student), Xu, Y. (PGR student), Tian, Y. (PGR student), Henderson, A. (Technical team), Morley, D. (Support team), Tallentire, J. (Support team), Smith, J. (Support team), Hawthornthwaite, S. (Support team), Carlson, J. (Support team) & Baniqued, P. D. (Researcher)
Project: Research

Cite this

@inproceedings{a8f97a4653024acd8a4ecfc7244ccc97,

title = "A Bilingual Cognitive Robot that Learns like a Toddler",

abstract = "We propose an embodied architecture featuring a developmental agent and a social robot for human-robot verbal engagement at preschool level. Initially, we trained the agent on bilingual acquisition and demonstrated its skill to appropriately detect the spoken content and automatically match the human user{\textquoteright}s language. Multilingual robot agents able to understand and communicate contexts in multiple tongues without loss of meaning in translations can greatly affect human cognitive and interaction capabilities. Furthermore, we teach the robot to gradually learn from experience new knowledge not specified at design time and autonomously use it for task-solving without being trained anew. The demonstrated system shows promising potential to becoming a low-cost social companion that can be easily integrated in the human world. ",

keywords = "brain-inspired architecture, developmental agent, cognitive system, embodiment, bilingual robot, multi-language development, humanlike learning",

author = "Ioanna Giorgi and Angelo Cangelosi and Masala, {Giovanni L.}",

year = "2020",

month = sep,

day = "18",

language = "English",

booktitle = "IEEE Symposium Series on Computational Intelligence (SSCI 2020)",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - A Bilingual Cognitive Robot that Learns like a Toddler

AU - Giorgi, Ioanna

AU - Cangelosi, Angelo

AU - Masala, Giovanni L.

PY - 2020/9/18

Y1 - 2020/9/18

N2 - We propose an embodied architecture featuring a developmental agent and a social robot for human-robot verbal engagement at preschool level. Initially, we trained the agent on bilingual acquisition and demonstrated its skill to appropriately detect the spoken content and automatically match the human user’s language. Multilingual robot agents able to understand and communicate contexts in multiple tongues without loss of meaning in translations can greatly affect human cognitive and interaction capabilities. Furthermore, we teach the robot to gradually learn from experience new knowledge not specified at design time and autonomously use it for task-solving without being trained anew. The demonstrated system shows promising potential to becoming a low-cost social companion that can be easily integrated in the human world.

AB - We propose an embodied architecture featuring a developmental agent and a social robot for human-robot verbal engagement at preschool level. Initially, we trained the agent on bilingual acquisition and demonstrated its skill to appropriately detect the spoken content and automatically match the human user’s language. Multilingual robot agents able to understand and communicate contexts in multiple tongues without loss of meaning in translations can greatly affect human cognitive and interaction capabilities. Furthermore, we teach the robot to gradually learn from experience new knowledge not specified at design time and autonomously use it for task-solving without being trained anew. The demonstrated system shows promising potential to becoming a low-cost social companion that can be easily integrated in the human world.

KW - brain-inspired architecture

KW - developmental agent

KW - cognitive system

KW - embodiment

KW - bilingual robot

KW - multi-language development

KW - humanlike learning

M3 - Conference contribution

BT - IEEE Symposium Series on Computational Intelligence (SSCI 2020)

PB - IEEE

ER -

A Bilingual Cognitive Robot that Learns like a Toddler

Abstract

Keywords

Fingerprint

Projects

RAI: Centre for Robotics and Artificial Intelligence

Cite this