Teaching a Machine to Read Maps with Deep Reinforcement Learning
arXiv:1711.07479
Abstract
The ability to use a 2D map to navigate a complex 3D environment is quite remarkable, and even difficult for many humans. Localization and navigation is also an important problem in domains such as robotics, and has recently become a focus of the deep reinforcement learning community. In this paper we teach a reinforcement learning agent to read a map in order to find the shortest way out of a random maze it has never seen before. Our system combines several state-of-the-art methods such as A3C and incorporates novel elements such as a recurrent localization cell. Our agent learns to localize itself based on 3D first person images and an approximate orientation angle. The agent generalizes well to bigger mazes, showing that it learned useful localization and navigation capabilities.
Paper accepted at 32nd AAAI Conference on Artificial Intelligence, AAAI 2018, New Orleans, Louisiana, USA
Cited by in corpus (5)
- Deep Reinforcement Learning: An Overview
- Learning to Navigate in Cities Without a Map
- Hierarchical Robot Navigation in Novel Environments using Rough 2-D Maps
- SafeRoute: Learning to Navigate Streets Safely in an Urban Environment
- DeepFoldit -- A Deep Reinforcement Learning Neural Network Folding Proteins