A Deep Reinforcement Learning Approach for Active SLAM

A Deep Reinforcement Learning Approach for Active SLAM

In this paper, we formulate the active SLAM paradigm in terms of model-free Deep Reinforcement Learning, embedding the traditional utility functions based on the Theory of Optimal Experimental Design in rewards, and therefore relaxing the intensive computations of classical approaches. We validate s...

Full description

Bibliographic Details
Main Authors: Julio A. Placed, José A. Castellanos
Format: Article
Language:English
Published: MDPI AG 2020-11-01
Series:Applied Sciences
Subjects:
active SLAM
robotic exploration
optimality criteria
deep reinforcement learning
deep Q-networks
Online Access:https://www.mdpi.com/2076-3417/10/23/8386

Internet

https://www.mdpi.com/2076-3417/10/23/8386

Similar Items