A semi-autonomous UAV platform for indoor remote operation with visual and haptic feedback

Paolo Stegagno, Massimo Basile, Heinrich H. Bulthoff, Antonio Franchi

Research output: Contribution to journalConference articlepeer-review

36 Citations (Scopus)

Abstract

We present the development of a semi-autonomous quadrotor UAV platform for indoor teleoperation using RGB-D technology as exceroceptive sensor. The platform integrates IMU and Dense Visual Odometry pose estimation in order to stabilize the UAV velocity and track the desired velocity commanded by a remote operator though an haptic interface. While being commanded, the quadrotor autonomously performs a persistent pan-scanning of the surrounding area in order to extend the intrinsically limited field of view. The RGB-D sensor is used also for collision-safe navigation using a probabilistically updated local obstacle map. In the operator visual feedback, pan-scanning movement is real time compensated by an IMU-based adaptive filtering algorithm that lets the operator perform the drive experience in a oscillation-free frame. An additional sensory channel for the operator is provided by the haptic feedback, which is based on the obstacle map and velocity tracking error in order to convey information about the environment and quadrotor state. The effectiveness of the platform is validated by means of experiments performed without the aid of any external positioning system.

Original languageEnglish
Article number6907419
Pages (from-to)3862-3869
Number of pages8
JournalProceedings - IEEE International Conference on Robotics and Automation
DOIs
Publication statusPublished - 2014 Sep 22
Event2014 IEEE International Conference on Robotics and Automation, ICRA 2014 - Hong Kong, China
Duration: 2014 May 312014 Jun 7

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A semi-autonomous UAV platform for indoor remote operation with visual and haptic feedback'. Together they form a unique fingerprint.

Cite this