How can a delivery robot navigate reliably to a destination in a new office building, with minimal prior information? To tackle this challenge, this paper introduces a two-level hierarchical approach, which integrates model-free deep learning and model-based path planning. At the low level, a neural-network motion controller, called the intention-net, is trained end-to-end to provide robust local navigation. The intention-net maps images from a single monocular camera and "intentions" directly to robot controls. At the high level, a path planner uses a crude map, e.g., a 2-D floor plan, to compute a path from the robot's current location to the goal. The planned path provides intentions to the intention-net. Preliminary experiments suggest that the learned motion controller is robust against perceptual uncertainty and by integrating with a path planner, it generalizes effectively to new environments and goals.
@article{arxiv.1710.05627,
title = {Intention-Net: Integrating Planning and Deep Learning for Goal-Directed Autonomous Navigation},
author = {Wei Gao and David Hsu and Wee Sun Lee and Shengmei Shen and Karthikk Subramanian},
journal= {arXiv preprint arXiv:1710.05627},
year = {2017}
}
Comments
Published in 1st Annual Conference on Robot Learning (CoRL 2017)