Skip to content

Deep Learning for Exploration and Recovery of Uncharted and Dynamic Targets from UAV-like Vision

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publication2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018)
Publisher or commissioning bodyInstitute of Electrical and Electronics Engineers (IEEE)
Pages1124-1131
Number of pages8
ISBN (Electronic)9781538680940
ISBN (Print)9781538680957
DOIs
DateAccepted/In press - 29 Jul 2018
DateE-pub ahead of print - 7 Jan 2019
DatePublished (current) - Jan 2019

Publication series

Name
ISSN (Print)2153-0858

Abstract

This paper discusses deep learning for solving static and dynamic search and recovery tasks – such as the retrieval of all instances of actively moving targets – based on partial-view Unmanned Aerial Vehicle (UAV)-like sensing. In particular, we demonstrate that abstracted tactic and strategic explorational agency can be implemented effectively via a single deep network that optimises in unity: the mapping of sensory inputs and positional history towards navigational actions. We propose a dual-stream classification paradigm that integrates one Convolutional Neural Network (CNN) for sensory processing with a second one for interpreting an evolving long-term map memory. In order to learn effective search behaviours given agent location and agent-centric sensory inputs, we train this design against 400k+ optimal navigational decision samples from each set of static and dynamic evolutions for different multi-target behaviour classes. We quantify recovery performance across an extensive range of scenarios; including probabilistic placement and dynamics, as well as fully random target walks and herd-inspired behaviours. Detailed results comparisons show that our design can outperform naïve, independent stream and off-the-shelf DRQN solutions. We conclude that the proposed dual-stream architecture can provide a unified, rationally motivated and effective architecture for solving online search tasks in dynamic, multi-target environments. With this paper we publish 3 3 Source code available at: https://data.bris.ac.uk/data and https://github.com/CWOA/GTRF key source code and associated models.

    Research areas

  • Navigation, robot sensing systems, task analysis, history, visualisation, vehicle dynamics, Reinforcement Learning

Download statistics

No data available

Documents

Documents

  • Full-text PDF (accepted author manuscript)

    Rights statement: This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at https://ieeexplore.ieee.org/document/8593751. Please refer to any applicable terms of use of the publisher.

    Accepted author manuscript, 2 MB, PDF-document

DOI

View research connections

Related faculties, schools or groups