Section: New Results

Visual servoing

Photometric moment-based visual servoing

Participants : Manikandan Bakthavatchalam, François Chaumette.

The goal of this work is to determine an adequate set of visual features to control the six degrees of freedom of a dynamic system. Thanks to a collaboration with Omar Tahri from Le2I in Le Creusot, we have been able to improve the results obtained previously with shifted moments for increasing the stability domain of visual servoing [24] .

Histogram-based visual servoing

Participants : Quentin Bateux, Eric Marchand.

Classically visual servoing considered the regulation in the image of a set of visual features (usually geometric features). Recently direct visual servoing schemes, such as photometric visual servoing, have been introduced in order to consider the image as a whole and thus avoid the extraction and the tracking of such geometric features. In this preliminary work, we propose a method to extend direct visual servoing approaches by using a global descriptor, namely intensity histograms, on the whole or multiple sub-sets of the images in order to achieve the control of a 6 degrees of freedom (DoF) robot.

Predictive visual servoing

Participants : Nicolas Cazy, Paolo Robuffo Giordano, François Chaumette.

This study is devoted to the application of predictive control to visual servoing. In a first step, we have developed and compared several predictive models that can be useful when some visual features leave the camera field of view or are lost because of occlusions [25] .

Visual servoing of cable-driven parallel robot

Participant : François Chaumette.

This study is realized in collaboration with Rémy Ramadour and Jean-Pierre Merlet from Coprin group at Inria Sophia Antipolis. Its goal is to adapt visual servoing techniques for cable-driven parallel robot in order to achieve accurate manipulation tasks [46] . This study is in the scope of the Inria large-scale initiative action PAL (see Section  8.2.6 ).


Participants : Le Cui, Eric Marchand.

We began a work, within the ANR P2N Nanorobust project (see Section  8.2.1 ), on the development of micro- and nano-manipulation within SEM (Scanning Electron Microscope). Our goal is to provide visual servoing techniques for positioning and manipulation tasks with a nanometer precision. This year, we focused on the characterisation of the projection model of a SEM along with the approach required for its calibration [26] . We then address the problem of 6 dof control using photometric feature under an optical microscope [27] . Finally, we focused on the definition of control law able to control the motion along the Z axes with a SEM microscope. Indeed, considering that a SEM is subject to parallel projection model, motion along this axis is not observable. We then address this problem using defocus information. An autofocus process has also been studied.

Audio-based control

Participants : Aly Magassouba, François Chaumette.

This study is not concerned with visual servoing, but to the application of the same principle of closed loop control schemes to audio sensors. It is made in collaboration with Nancy Bertin from Panama group at Inria Rennes-Bretagne Atlantique. In a first step, we have determined the analytical form of the interaction matrix of audio features based on the time difference of arrival on two microphones. From this modeling step, we have determined the different virtual linkages that can be realized in function of the number and configuration of sources.