Visual identification of biological motion for underwater human–robot interaction

Junaed Sattar, Gregory Dudek

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We present an algorithm for underwater robots to visually detect and track human motion. Our objective is to enable human–robot interaction by allowing a robot to follow behind a human moving in (up to) six degrees of freedom. In particular, we have developed a system to allow a robot to detect, track and follow a scuba diver by using frequency-domain detection of biological motion patterns. The motion of biological entities is characterized by combinations of periodic motions which are inherently distinctive. This is especially true of human swimmers. By using the frequency-space response of spatial signals over a number of video frames, we attempt to identify signatures pertaining to biological motion. This technique is applied to track scuba divers in underwater domains, typically with the robot swimming behind the diver. The algorithm is able to detect a range of motions, which includes motion directly away from or towards the camera. Once detected, the motion of the diver relative to the vehicle is then tracked using an Unscented Kalman Filter, an approach for non-linear estimation. The efficiency of our approach makes it attractive for real-time applications on-board our underwater vehicle, and in future applications we intend to track scuba divers in real-time with the robot. The paper presents an algorithmic overview of our approach, together with experimental evaluation based on underwater video footage.

Original languageEnglish (US)
Pages (from-to)111-124
Number of pages14
JournalAutonomous Robots
Volume42
Issue number1
DOIs
StatePublished - Jan 1 2018

Fingerprint

Robots
Kalman filters
Cameras

Keywords

  • Diver tracking
  • Human–robot collaboration
  • Underwater robotics
  • Visual servoing
  • Visual tracking

Cite this

Visual identification of biological motion for underwater human–robot interaction. / Sattar, Junaed; Dudek, Gregory.

In: Autonomous Robots, Vol. 42, No. 1, 01.01.2018, p. 111-124.

Research output: Contribution to journalArticle

@article{d207a2497a4345afb7ea4f564113c38d,
title = "Visual identification of biological motion for underwater human–robot interaction",
abstract = "We present an algorithm for underwater robots to visually detect and track human motion. Our objective is to enable human–robot interaction by allowing a robot to follow behind a human moving in (up to) six degrees of freedom. In particular, we have developed a system to allow a robot to detect, track and follow a scuba diver by using frequency-domain detection of biological motion patterns. The motion of biological entities is characterized by combinations of periodic motions which are inherently distinctive. This is especially true of human swimmers. By using the frequency-space response of spatial signals over a number of video frames, we attempt to identify signatures pertaining to biological motion. This technique is applied to track scuba divers in underwater domains, typically with the robot swimming behind the diver. The algorithm is able to detect a range of motions, which includes motion directly away from or towards the camera. Once detected, the motion of the diver relative to the vehicle is then tracked using an Unscented Kalman Filter, an approach for non-linear estimation. The efficiency of our approach makes it attractive for real-time applications on-board our underwater vehicle, and in future applications we intend to track scuba divers in real-time with the robot. The paper presents an algorithmic overview of our approach, together with experimental evaluation based on underwater video footage.",
keywords = "Diver tracking, Human–robot collaboration, Underwater robotics, Visual servoing, Visual tracking",
author = "Junaed Sattar and Gregory Dudek",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/s10514-017-9644-y",
language = "English (US)",
volume = "42",
pages = "111--124",
journal = "Autonomous Robots",
issn = "0929-5593",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Visual identification of biological motion for underwater human–robot interaction

AU - Sattar, Junaed

AU - Dudek, Gregory

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We present an algorithm for underwater robots to visually detect and track human motion. Our objective is to enable human–robot interaction by allowing a robot to follow behind a human moving in (up to) six degrees of freedom. In particular, we have developed a system to allow a robot to detect, track and follow a scuba diver by using frequency-domain detection of biological motion patterns. The motion of biological entities is characterized by combinations of periodic motions which are inherently distinctive. This is especially true of human swimmers. By using the frequency-space response of spatial signals over a number of video frames, we attempt to identify signatures pertaining to biological motion. This technique is applied to track scuba divers in underwater domains, typically with the robot swimming behind the diver. The algorithm is able to detect a range of motions, which includes motion directly away from or towards the camera. Once detected, the motion of the diver relative to the vehicle is then tracked using an Unscented Kalman Filter, an approach for non-linear estimation. The efficiency of our approach makes it attractive for real-time applications on-board our underwater vehicle, and in future applications we intend to track scuba divers in real-time with the robot. The paper presents an algorithmic overview of our approach, together with experimental evaluation based on underwater video footage.

AB - We present an algorithm for underwater robots to visually detect and track human motion. Our objective is to enable human–robot interaction by allowing a robot to follow behind a human moving in (up to) six degrees of freedom. In particular, we have developed a system to allow a robot to detect, track and follow a scuba diver by using frequency-domain detection of biological motion patterns. The motion of biological entities is characterized by combinations of periodic motions which are inherently distinctive. This is especially true of human swimmers. By using the frequency-space response of spatial signals over a number of video frames, we attempt to identify signatures pertaining to biological motion. This technique is applied to track scuba divers in underwater domains, typically with the robot swimming behind the diver. The algorithm is able to detect a range of motions, which includes motion directly away from or towards the camera. Once detected, the motion of the diver relative to the vehicle is then tracked using an Unscented Kalman Filter, an approach for non-linear estimation. The efficiency of our approach makes it attractive for real-time applications on-board our underwater vehicle, and in future applications we intend to track scuba divers in real-time with the robot. The paper presents an algorithmic overview of our approach, together with experimental evaluation based on underwater video footage.

KW - Diver tracking

KW - Human–robot collaboration

KW - Underwater robotics

KW - Visual servoing

KW - Visual tracking

UR - http://www.scopus.com/inward/record.url?scp=85020720591&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020720591&partnerID=8YFLogxK

U2 - 10.1007/s10514-017-9644-y

DO - 10.1007/s10514-017-9644-y

M3 - Article

VL - 42

SP - 111

EP - 124

JO - Autonomous Robots

JF - Autonomous Robots

SN - 0929-5593

IS - 1

ER -