Recent Advances in Robot Learning from Demonstration

In the context of robotics and automation, learning from demonstrations (LfD) is the paradigm in which robots acquire new skills by learning to imitate an expert. The choice of LfD over other robot learning methods is compelling when ideal behavior can be neither easily scripted, as done in traditional robot programming, nor easily defined as an optimization problem using a reward function, but can be demonstrated. While there have been multiple surveys of the field in the past, there is a need for a new survey given the considerable growth in the number of publications in recent years. This survey aims at offering an overview of the collection of machine learning methods used to enable a robot to learn from and imitate a teacher. We focus on recent advancements in the field, as well as present an updated taxonomy and characterization of existing methods. We also discuss mature and emerging application areas for LfD, and highlight the significant challenges that remain to be overcome both in theory and practice.

 @article{ravichandar2019lfd,
  title = {Recent Advances in Robot Learning from Demonstration},
  author = {Ravichandar, Harish and Polydoros, Athanasios S. and Chernova, Sonia and Billard, Aude},
  journal = {Annual Review of Control, Robotics, and Autonomous Systems},
  pages = {In Press},
  year = {2019},
  publisher = {Annual Reviews},
  pdflink1 = {https://pdfs.semanticscholar.org/f441/e637980a8b427474dbdc0141f38dd78bb831.pdf},
  public = {yes}
}