Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues

Jose Baleia; Pedro Santana; Jose Barata

doi:10.1109/ICARSC.2014.6849777

Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues

Jose Baleia, Pedro Santana, Jose Barata

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Citations (Scopus)

Abstract

This paper presents a ground vehicle capable of exploiting haptic cues to learn navigation affordances from depth cues. A simple pan-tilt telescopic antenna and a Kinect sensor, both fitted to the robot's body frame, provide the required haptic and depth sensory feedback, respectively. With the antenna, the robot determines whether an object is traversable by the robot. Then, the interaction outcome is associated to the object's depth-based descriptor. Later on, the robot to predict if a newly observed object is traversable just by inspecting its depth-based appearance uses this acquired knowledge. A set of field trials show the ability of the to robot progressively learn which elements of the environment are traversable.

Original language	English
Title of host publication	2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC
Editors	N Lau, AP Moreira, R Ventura, BM Faria
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	146-151
Number of pages	6
DOIs	https://doi.org/10.1109/ICARSC.2014.6849777
Publication status	Published - 2014
Event	IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC) - Espinho, Portugal Duration: 14 May 2014 → 15 May 2014

Conference

Conference	IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)
Country/Territory	Portugal
City	Espinho
Period	14/05/14 → 15/05/14

Keywords

affordances
autonomous robots
depth sensing
robotic antenna
self-supervised learning
terrain assessment

Access to Document

10.1109/ICARSC.2014.6849777

Cite this

Baleia, J., Santana, P., & Barata, J. (2014). Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues. In N. Lau, AP. Moreira, R. Ventura, & BM. Faria (Eds.), 2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC (pp. 146-151). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ICARSC.2014.6849777

@inproceedings{ce2f0c6889764f9e951db10b99686841,

title = "Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues",

abstract = "This paper presents a ground vehicle capable of exploiting haptic cues to learn navigation affordances from depth cues. A simple pan-tilt telescopic antenna and a Kinect sensor, both fitted to the robot's body frame, provide the required haptic and depth sensory feedback, respectively. With the antenna, the robot determines whether an object is traversable by the robot. Then, the interaction outcome is associated to the object's depth-based descriptor. Later on, the robot to predict if a newly observed object is traversable just by inspecting its depth-based appearance uses this acquired knowledge. A set of field trials show the ability of the to robot progressively learn which elements of the environment are traversable.",

keywords = "affordances, autonomous robots, depth sensing, robotic antenna, self-supervised learning, terrain assessment",

author = "Jose Baleia and Pedro Santana and Jose Barata",

note = "This work was co-funded by ROBOSAMPLER project (LISBOA-01-0202-FEDER-024961) and by the CTS multiannual funding, through the PIDDAC Program funds.; IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC) ; Conference date: 14-05-2014 Through 15-05-2014",

year = "2014",

doi = "10.1109/ICARSC.2014.6849777",

language = "English",

pages = "146--151",

editor = "N Lau and AP Moreira and R Ventura and BM Faria",

booktitle = "2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

Baleia, J, Santana, P & Barata, J 2014, Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues. in N Lau, AP Moreira, R Ventura & BM Faria (eds), 2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC . Institute of Electrical and Electronics Engineers (IEEE), pp. 146-151, IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), Espinho, Portugal, 14/05/14. https://doi.org/10.1109/ICARSC.2014.6849777

Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues. / Baleia, Jose; Santana, Pedro; Barata, Jose.
2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC . ed. / N Lau; AP Moreira; R Ventura; BM Faria. Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 146-151.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues

AU - Baleia, Jose

AU - Santana, Pedro

AU - Barata, Jose

N1 - This work was co-funded by ROBOSAMPLER project (LISBOA-01-0202-FEDER-024961) and by the CTS multiannual funding, through the PIDDAC Program funds.

PY - 2014

Y1 - 2014

N2 - This paper presents a ground vehicle capable of exploiting haptic cues to learn navigation affordances from depth cues. A simple pan-tilt telescopic antenna and a Kinect sensor, both fitted to the robot's body frame, provide the required haptic and depth sensory feedback, respectively. With the antenna, the robot determines whether an object is traversable by the robot. Then, the interaction outcome is associated to the object's depth-based descriptor. Later on, the robot to predict if a newly observed object is traversable just by inspecting its depth-based appearance uses this acquired knowledge. A set of field trials show the ability of the to robot progressively learn which elements of the environment are traversable.

AB - This paper presents a ground vehicle capable of exploiting haptic cues to learn navigation affordances from depth cues. A simple pan-tilt telescopic antenna and a Kinect sensor, both fitted to the robot's body frame, provide the required haptic and depth sensory feedback, respectively. With the antenna, the robot determines whether an object is traversable by the robot. Then, the interaction outcome is associated to the object's depth-based descriptor. Later on, the robot to predict if a newly observed object is traversable just by inspecting its depth-based appearance uses this acquired knowledge. A set of field trials show the ability of the to robot progressively learn which elements of the environment are traversable.

KW - affordances

KW - autonomous robots

KW - depth sensing

KW - robotic antenna

KW - self-supervised learning

KW - terrain assessment

U2 - 10.1109/ICARSC.2014.6849777

DO - 10.1109/ICARSC.2014.6849777

M3 - Conference contribution

SP - 146

EP - 151

BT - 2014 IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC

A2 - Lau, N

A2 - Moreira, AP

A2 - Ventura, R

A2 - Faria, BM

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)

Y2 - 14 May 2014 through 15 May 2014

ER -

Self-Supervised Learning of Depth-Based Navigation Affordances from Haptic Cues

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this