Learning Human-Aware Path Planning with Fully Convolutional Networks

N. Perez-Higueras, Fernando Caballero, and Luis Merino. Learning Human-Aware Path Planning with Fully Convolutional Networks. In Proceedings of the IEEE International Conference on Robotics and Automation, ICRA, pp. 5897–5902, 2018.

Download

[PDF] [html] [code] [dataset] 

Abstract

This work presents an approach to learn path planning for robot social navigation by demonstration. We make use of Fully Convolutional Neural Networks (FCNs) to learn from expert's path demonstrations a map that marks a feasible path to the goal as a classification problem. The use of FCNs allows us to overcome the problem of manually designing/identifying the cost-map and relevant features for the task of robot navigation. The method makes use of optimal Rapidly-exploring Random Tree planner (RRT*) to overcome eventual errors in the path prediction; the FCNs prediction is used as cost-map and also to partially bias the sampling of the configuration space, leading the planner to behave similarly to the learned expert behavior. The approach is evaluated in experiments with real trajectories and compared with Inverse Reinforcement Learning algorithms that use RRT* as underlying planner.

BibTeX Entry

@INPROCEEDINGS{icra18fcnrrt,
  author = {N. Perez-Higueras and Fernando Caballero and Luis Merino},
  title = {{Learning Human-Aware Path Planning with Fully Convolutional Networks}},
  booktitle = ICRA,
  year = {2018},
  pages = {5897--5902},
  doi ={10.1109/ICRA.2018.8460851},
  abstract={This work presents an approach to learn path planning for robot social navigation by demonstration. We make use of Fully Convolutional Neural Networks (FCNs) to learn from expert's path demonstrations a map that marks a feasible path to the goal as a classification problem. The use of FCNs allows us to overcome the problem of manually designing/identifying the cost-map and relevant features for the task of robot navigation. The method makes use of optimal Rapidly-exploring Random Tree planner (RRT*) to overcome eventual errors in the path prediction; the FCNs prediction is used as cost-map and also to partially bias the sampling of the configuration space, leading the planner to behave similarly to the learned expert behavior. The approach is evaluated in experiments with real trajectories and compared with Inverse Reinforcement Learning algorithms that use RRT* as underlying planner.},
}

Generated by bib2html.pl (written by Patrick Riley ) on Mon Jan 09, 2023 18:40:36