Application of Neural Networks to Image-Based Control of Robot Arms

Venaille, Christophe; Wells, Gordon; Torras, Carme

doi:10.1016/s1474-6670(17)46120-9

Cited by 2 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our previous work demonstrates how global image encodings such as Fourier descriptors [34] and geometric moments [35] may be applied to complex, real images to position a 6-dof industrial robot. Although the first results obtained with this prototype were very encouraging, the precision attained was not yet as desired for some of the degrees of freedom, and varied considerably among them.…”

Section: Image Featuresmentioning

confidence: 99%

Untitled

Wells

Torras

2001

Journal of Intelligent and Robotic Systems

Self Cite

View full text Add to dashboard Cite

The development of any robotics application relying on visual information always raises the key question of what image features would be most informative about the motion to be performed. In this paper, we address this question in the context of visual robot positioning, where a neural network is used to learn the mapping between image features and robot movements, and global image descriptors are preferred to local geometric features. Using a statistical measure of variable interdependence called Mutual Information, subsets of image features most relevant for determining pose variations along each of the six degrees of freedom (dof's) of camera motion are selected. Four families of global features are considered: geometric moments, eigenfeatures, Local Feature Analysis vectors, and a novel feature called Pose-Image Covariance vectors. The experimental results described show the quantitative and qualitative benefits of performing this feature selection prior to training the neural network: Less network inputs are needed, thus considerably shortening training times; the dof's that would yield larger errors can be determined beforehand, so that more informative features can be sought; the order of the features selected for each dof often accepts an intuitive explanation, which in turn helps to provide insights for devising features tailored to each dof.

show abstract

Section: Image Featuresmentioning

confidence: 99%