Saad Khan's Publications

Sorted by DateClassified by Publication TypeClassified by Research Category

Towards learning movement in dense crowds for a socially-aware mobile robot

S.A. Khan, S. Arif, and L. Bölöni. Towards learning movement in dense crowds for a socially-aware mobile robot. In Proc. of Adaptive Learning Agents workshop at AAMAS-2014, May 2014.

Download

[PDF] 

Abstract

Robots moving in a crowd occasionally reach situations where they need to decide whether to give way to a human or not, a situation we call a micro-conflict and model with a two player game. We collect data from a robot controlled by a human operator and use three different supervised learning algorithms (random forest, SVM and neuro-evolution) to create a decision maker module which imitates the human operator's behavior in micro-conflicts. Results show that the neuro-evolution based decision-maker gives the best performance under scenarios with various crowd density and urgency. In addition, we found that the neuro-evolution method generalizes better to environments very different from those in the training set.

BibTeX

@inproceedings{SKhan-2014-ALA,
   title = "Towards learning movement in dense crowds for a socially-aware mobile robot",
   author = "S.A. Khan and S. Arif and  L. B{\"o}l{\"o}ni",
   booktitle = "Proc. of Adaptive Learning Agents workshop at AAMAS-2014",
   year = "2014",
   bib2html_dl_pdf =
   {http://www.eecs.ucf.edu/~skhan/Publications/Download/SKhan-2014-ALA.pdf},
   month = "May",
   bib2html_pubtype = {Refereed Conference},
   bib2html_rescat = {Autonomous Agents},
   abstract = {
   Robots moving in a crowd occasionally reach situations where they need to decide whether to give way to a human or not, a situation we call a {\em micro-conflict} and model with a two player game. We collect data from a robot controlled by a human operator and use three different supervised learning algorithms (random forest, SVM and neuro-evolution) to create a decision maker module which imitates the human operator's behavior in micro-conflicts. Results show that the neuro-evolution based decision-maker gives the best performance under scenarios with various crowd density and urgency. In addition, we found that the neuro-evolution method generalizes better to environments very different from those in the training set.
},
}

Generated by bib2html.pl (written by Patrick Riley, Saad Khan ) on Thu Jul 10, 2014 15:43:58