&lt;title&gt;Neural-network-based vision processing for autonomous robot guidance&lt;/title&gt;

Pomerleau, Dean A.

doi:10.1117/12.44950

Cited by 15 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Fortunately, there are a number of tasks, particularly in robotics, that require determining the precise position of features in the environment. In addition to guiding the SM 2, these techniques have been successfully applied to steering an autonomous automobile at speeds of up to 55 mph (Pomerleau, 1991b). Work is also in progress in using these connectionist methods to guide the landing of an unmanned flying vehicle (Davis & Stentz, 1992).…”

Section: Discussionmentioning

confidence: 99%

“…Given an image coming from the camera at the gripper, the vision system needs to provide the two-dimensional displacement of the gripper screw relative to the anchor hole. To accomplish this mapping, a neural network architecture very similar to that in Pomerleau (1991aPomerleau ( , 1991b is used. Specifically, the input layer consists of a 24 x 20 two-dimensional "retina" that receives input from the gripper's video camera.…”

Section: Network Architecturementioning

confidence: 99%

“…The article focuses on the use of artificial neural networks for the precise control of the Self Mobile Space Manipulator (SM2), a robot designed to walk on the external structure of the space station Freedom (Brown, Friedman, & Kanade, 1990;Ueno, Xu, & Brown, 1990). Application of similar neural network techniques for autonomous robot driving can be found in Pomerleau (1991aPomerleau ( , 1991b).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Neural network-based vision for precise control of a walking robot

Pomerleau

1994

Mach Learn

Self Cite

View full text Add to dashboard Cite

This article describes a connectionist vision system for the precise control of a robot designed to walk on the exterior of the space station. The network learns to use video camera input to determine the displacement of the robot's gripper relative to a hole in which the gripper must be inserted. Once trained, the network's output is used to control the robot, with a resulting factor of five fewer missed gripper insertions than occur when the robot walks without sensor feedback. The neural network visual feedback techniques described could also be applied in domains such as manufacturing, where precise robot positioning is required in an uncertain environment.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Network Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Neural network-based vision for precise control of a walking robot

Pomerleau

1994

Mach Learn

Self Cite

View full text Add to dashboard Cite

show abstract

“…Fortunately, there are a number of tasks, particularly in robotics, that require determining the precise position of features in the environment. In addition to guiding the SM2, these techniques have been successfully applied to steering an autonomous automobile at speeds of up to 55 mph (Pomerleau, 1991b). Work is also in progress in using these connectionist methods to guide the landing of an unmanned flying vehicle (Davis & Stentz, 1992).…”

Section: Discussionmentioning

confidence: 99%

“…To accomplish this mapping, a neural network architecture very similar to that in Pomerleau (1991aPomerleau ( , 1991b) is used. Specifically, the input layer consists of a 24 x 20 two-dimensional "retina" that receives input from the gripper's video camera.…”

Section: Network Architecturementioning

confidence: 99%

Untitled

Pomerleau

1994

Machine Learning

Self Cite

View full text Add to dashboard Cite

Abstract. This article describes a connectionist vision system for the precise control of a robot designed to walk on the exterior of the space station. The network learns to use video camera input to determine the displacement of the robot's gripper relative to a hole in which the gripper must be inserted. Once trained, the network's output is used to control the robot, with a resulting factor of five fewer missed gripper insertions than occur when the robot walks without sensor feedback. The neural network visual feedback techniques described could also be applied in domains such as manufacturing, where precise robot positioning is required in an uncertain environment.

show abstract

Neural Systems for Control

Omidvar¹,

Elliott²

1997

View full text Add to dashboard Cite

<title>Neural-network-based vision processing for autonomous robot guidance</title>

Cited by 15 publications

References 7 publications

Neural network-based vision for precise control of a walking robot

Neural network-based vision for precise control of a walking robot

Untitled

Neural Systems for Control

Contact Info

Product

Resources

About