Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


Variety Wins: Soccer-Playing Robots and Infant Walking

Ori Ossmy, Justine E. Hoch, Patrick MacAlpine, Shohan Hasan, Peter Stone, and Karen E. Adolph. Variety Wins: Soccer-Playing Robots and Infant Walking. Frontiers in Neurorobotics, 12:19, 2018.
Available from the publisher's webpage

Download

[PDF]2.9MB  

Abstract

Although both infancy and artificial intelligence (AI) researchers are interested in developing systems that produce adaptive, functional behavior, the two disciplines rarely capitalize on their complementary expertise. Here, we used soccer-playing robots to test a central question about the development of infant walking. During natural activity, infants' locomotor paths are immensely varied. They walk along curved, multi-directional paths with frequent starts and stops. Is the variability observed in spontaneous infant walking a "feature" or a "bug"? In other words, is variability beneficial for functional walking performance? To address this question, we trained soccer-playing robots on walking paths generated by infants during free play and tested them in simulated games of "RoboCup." In Tournament 1, we compared the functional performance of a simulated robot soccer team trained on infants' natural paths with teams trained on less varied, geometric paths—straight lines, circles, and squares. Across 1000 head-to-head simulated soccer matches, the infant-trained team consistently beat all teams trained with less varied walking paths. In Tournament 2, we compared teams trained on different clusters of infant walking paths. The team trained with the most varied combination of path shape, step direction, number of steps, and number of starts and stops outperformed teams trained with less varied paths. This evidence indicates that variety is a crucial, functional feature supporting functional walking performance. More generally, we propose that robotics provides a fruitful avenue for testing hypotheses about infant development; reciprocally, behavioral observations of infant behavior may inform research on artificial intelligence.

BibTeX Entry

@article{FNBOT18-Ossmy,  
  author={Ori Ossmy and Justine E. Hoch and Patrick MacAlpine and Shohan Hasan and Peter Stone and Karen E. Adolph}, 
  title={Variety Wins: Soccer-Playing Robots and Infant Walking},
  journal={Frontiers in Neurorobotics},
  volume={12},
  pages={19},    	
  year={2018},
  url={https://www.frontiersin.org/article/10.3389/fnbot.2018.00019},
  doi={10.3389/fnbot.2018.00019},
  issn={1662-5218},
  wwwnote = {Available from the <a href="https://www.frontiersin.org/article/10.3389/fnbot.2018.00019">publisher's webpage</a>},
  abstract={Although both infancy and artificial intelligence (AI) researchers are 
interested in developing systems that produce adaptive, functional behavior, the 
two disciplines rarely capitalize on their complementary expertise. Here, we 
used soccer-playing robots to test a central question about the development of 
infant walking. During natural activity, infants' locomotor paths are immensely 
varied. They walk along curved, multi-directional paths with frequent starts and 
stops. Is the variability observed in spontaneous infant walking a "feature" or 
a "bug"? In other words, is variability beneficial for functional walking 
performance? To address this question, we trained soccer-playing robots on 
walking paths generated by infants during free play and tested them in simulated 
games of "RoboCup." In Tournament 1, we compared the functional performance of a 
simulated robot soccer team trained on infants' natural paths with teams trained 
on less varied, geometric paths—straight lines, circles, and squares. Across 
1000 head-to-head simulated soccer matches, the infant-trained team consistently 
beat all teams trained with less varied walking paths. In Tournament 2, we 
compared teams trained on different clusters of infant walking paths. The team 
trained with the most varied combination of path shape, step direction, number 
of steps, and number of starts and stops outperformed teams trained with less 
varied paths. This evidence indicates that variety is a crucial, functional 
feature supporting functional walking performance. More generally, we propose 
that robotics provides a fruitful avenue for testing hypotheses about infant 
development; reciprocally, behavioral observations of infant behavior may inform 
research on artificial intelligence.}
}

Generated by bib2html.pl (written by Patrick Riley ) on Tue Nov 24, 2020 14:00:04