Bodily aware soft robots: integration of proprioceptive and exteroceptive sensors

Gabor Soter; Jonathan Rossiter; Andrew Conn; Helmut Hauser

Bodily aware soft robots: integration of proprioceptive and exteroceptive sensors

Gabor Soter, Jonathan Rossiter, Andrew Conn, Helmut Hauser

Research output: Contribution to conference › Conference Poster › peer-review

348 Downloads (Pure)

Abstract

Being aware of our body has great importance in our everyday life. This is the reason why we know how to move in a dark room or to grasp a complex object. These skills are important for robots as well, however, robotic bodily awareness is still not solved. In this paper we present a novel method to implement bodily awareness into soft robots by the integration of exteroceptive and proprioceptive sensors. We use a combination of a stacked convolutional autoencoder and a recurrent neural network to map internal sensory signals to visual information. As a result, the simulated soft robot can learn to imagine its motion even when its visual sensor is not available.

Original language	English
Publication status	Unpublished - 15 Nov 2017
Event	Conference on Robot Learning 2017 - Mountain View, United States Duration: 13 Nov 2017 → 15 Nov 2017 Conference number: 1st

Conference

Conference	Conference on Robot Learning 2017
Abbreviated title	CoRL 2017
Country/Territory	United States
City	Mountain View
Period	13/11/17 → 15/11/17

Research Groups and Themes

Tactile Action Perception

Keywords

soft robots
bodily awareness
proprioception

Access to Document

Full-text PDF (accepted author manuscript)Accepted author manuscript, 5.79 MB

Handle.net

Persistent link

Cite this

@conference{9b9dbc4820224fbc9067b5e8c185cee1,

title = "Bodily aware soft robots: integration of proprioceptive and exteroceptive sensors",

abstract = "Being aware of our body has great importance in our everyday life. This is the reason why we know how to move in a dark room or to grasp a complex object. These skills are important for robots as well, however, robotic bodily awareness is still not solved. In this paper we present a novel method to implement bodily awareness into soft robots by the integration of exteroceptive and proprioceptive sensors. We use a combination of a stacked convolutional autoencoder and a recurrent neural network to map internal sensory signals to visual information. As a result, the simulated soft robot can learn to imagine its motion even when its visual sensor is not available.",

keywords = "soft robots, bodily awareness, proprioception",

author = "Gabor Soter and Jonathan Rossiter and Andrew Conn and Helmut Hauser",

year = "2017",

month = nov,

day = "15",

language = "English",

note = "Conference on Robot Learning 2017, CoRL 2017 ; Conference date: 13-11-2017 Through 15-11-2017",

}

TY - CONF

T1 - Bodily aware soft robots

T2 - Conference on Robot Learning 2017

AU - Soter, Gabor

AU - Rossiter, Jonathan

AU - Conn, Andrew

AU - Hauser, Helmut

N1 - Conference code: 1st

PY - 2017/11/15

Y1 - 2017/11/15

N2 - Being aware of our body has great importance in our everyday life. This is the reason why we know how to move in a dark room or to grasp a complex object. These skills are important for robots as well, however, robotic bodily awareness is still not solved. In this paper we present a novel method to implement bodily awareness into soft robots by the integration of exteroceptive and proprioceptive sensors. We use a combination of a stacked convolutional autoencoder and a recurrent neural network to map internal sensory signals to visual information. As a result, the simulated soft robot can learn to imagine its motion even when its visual sensor is not available.

AB - Being aware of our body has great importance in our everyday life. This is the reason why we know how to move in a dark room or to grasp a complex object. These skills are important for robots as well, however, robotic bodily awareness is still not solved. In this paper we present a novel method to implement bodily awareness into soft robots by the integration of exteroceptive and proprioceptive sensors. We use a combination of a stacked convolutional autoencoder and a recurrent neural network to map internal sensory signals to visual information. As a result, the simulated soft robot can learn to imagine its motion even when its visual sensor is not available.

KW - soft robots

KW - bodily awareness

KW - proprioception

M3 - Conference Poster

Y2 - 13 November 2017 through 15 November 2017

ER -

Bodily aware soft robots: integration of proprioceptive and exteroceptive sensors

Abstract

Conference

Research Groups and Themes

Keywords

Access to Document

Handle.net

Fingerprint

Cite this