Areas of interest
- Computer vision
- Image processing
- Visual servoing
- Surgical Robotics
- Robotics
Quick Access
- Research activity within Tecnalia Research & Innovation (since 2008)
- Research activity within CEA (2006-2007)
- Research activity at IRISA (2001-2006)
Research activity within Tecnalia Research & Innovation (since 2008)
Surgical Robotics
This activity was funded by the European project STIFF-FLOP
This project was motivated by the current limitations of modern laparoscopic and robot-assisted surgical systems that are due to restricted access through Trocar ports, lack of haptic feedback, and difficulties with rigid robot tools operating inside a confined space filled with organs.
Taking inspiration from biological manipulators, like the octopus arm, a soft robotic arm was designed for enabling access, through bending, to areas non directly accessible with standard rigid tools.
In the context of the project, we developed:
- a sensorised benchmarking platform for analyzing the interaction forces in between the STIFF-FLOP arm and objects with variable stiffness.
- a generic position-based controller, able to deduce the appropriate configuration of the arm modules to reach a desired tip pose. The modular inverse kinematics model enables considering different features for characterizing the modules configuration (chamber length, constant curvature, position, pressure)
- an extension of the inverse kinematics to include in the process the estimation of the appropriate pose of the standard robotic arm holding the STIFF-FLOP flexible structure, while maintaining the single point insertion constraint.
- an extension of the inverse kinematics, through secondary tasks, for reducing the extension of the modules, or limiting the contacts detected with the external world.
Work in collaboration with Fabrice Morin, Asier Fernandez, Julius Klein and Alfonso Dominguez.
Human-robot physical interaction
This activity was funded by the Spanish Plan Nacional Fluent, and the European Project CogLaboration .
In the context of human robot collaboration, we are developping some control laws dedicated to human robot object handovers. These control laws are strongly inspired by the Dynamic Movement Primitives that permit to reproduce a reference motion pattern while enabling to adjust online the targeted location.
Work in collaboration with Miguel Prada, Irati Rasines and Leire Martinez.
Assistive robotics
It is considered that assistive robots could provide elderly people with a physical and cognitive support that would permit to live longer independently in their home. In that context we have been contributing to different projects aiming at designing some mobile robot assistants providing different kind of supports to the elderly. Our contribution in that line mainly relies on the preparation of robotic enablers, in which the interaction with the mobile platform is abstracted for facilitating the use of the robot within the higher-level mechanisms.
Vision-based audio-guide
In collaboration with the company Auralia, we have been working on a vision-based audio-guide, in which the paintings are automatically detected by the audio-guide through image processing. That permits to automatically launch the audio description related to the observed painting. A pose estimation is also used to enrich the audio description with some 3D spatial resolution.
The details of this work are unfortunately confidential.
Vision for smart illumination
It is demonstrated that a good management of city and road illumination could permit to drastically reduce the related energy consumption. In that line, in collaboration with the Luix company, we have been working on the design of a vision-based presence detection technique, that would permit to adjust accordingly the intensity of the street lamp leds.
The details of this work are unfortunately confidential.
Vision-based wheelchair control, headMove (2008-2009)
The objective of this work is to give to persons with strong physical disabilities the access to electric wheelchairs that would permit them to increase their mobility and independance. Persons with spinal cord injury have their mobility capacities restricted to the head, which makes the control of traditional wheelchairs impossible. We proposed a control mechanism based on the visual tracking of the head motion.
The details of this work are unfortunately confidential.
Research activity within CEA (2006-2007)
This research is performed within two projects. The European project ITEA ANSO gathers about 15 partners working together for designing technical solutions towards a domotic environement. AVISO is a French project, in collaboration with the association Approche (assistance to disable people). This partnership will allow the CEA to evaluate with real end-users their scientific productions within several medical centers.
Towards Robotic Assistance for injured people
In few words, the CEA research concerns the design of a robotic arm able to grasp objects within a apartment-like environment. A stereoscopic rig fixed onto the gripper is used to control the arm. The end-user defines onto a display an object to grasp (by simply defining a box around it), and a visual servoing task is performed to go and fetch the defined object.
Contribution to the project
Within this context, my work was to :
- develop a stereo tracking algorithm. The main originality is that no object model is needed.
- design a control module (based on visual servoing principles) to command the arm.
The development of such elements enabled me to define several scientific libraries dealing with :
- algebra
- image grabbing
- camera calibration
- visual servoing
- feature extraction (harris points) and tracking (differential correlation), template matching (exhaustive correlation)
- pose estimation
Naturally, these libraries are now easily usable by the rest of the laboratory
The autonomous mobile robotic platform SAM
Within the ANSO project, all these physical and logicial elements are embedded into a mobile platform (developped by Neobotix), which enables the user to perform a grasping wherever in his apartment. This original application has the following skills:
- IHM designed for non specialists end-users, minimizing the number of interactions,
- Client /server mode enabling to interact with the mobile system, through a DPWS stack onto a wifi network,
- Action scheduling and monitoring controlled by a generic scenario interpretor layer,
- Autonomous mobile robot navigation (developed by Neobotix),
- Automatic object grasping (after a simple definition by the user).
Work in collaboration with Christophe Leroux, Gerard Chalubert, Martine Guerrand, Aline Chansavang, Céline Teulière & Mathieu le Cam.
Research activity at IRISA (2001-2006)
Research performed at IRISA where involved within French national projects Predit Mobivip and Robea Bodega. During these years in Bretagne, I was member of the research projects Lagadic and Texmex.
Robotic navigation
This work concerns a topological approach of vision-based navigation. Within this framework, the navigation space is described by an image data-base acquired off-line. The localization of the robotic system is performed by using image retrieval methods. Indeed, as long as the current position of the robotic system is defined by the image acquired by the en-boarded camera, the localization is nothing but the search of the views from the database that are the most similar in term of contents. The database is organized in a connectivity graph which gives a nice solution for defining a path between two images respectively describing the initial and desired positions, by performing a shortest path search. Original visual servoing methods are then proposed to control in-line the robot motions, by comparing the current view and the image path.
This work has been integrated onto the robotic car Cycab that
belongs to IRISA. These experiments in real environments have shown
the efficiency of the approach, and also its robustness towards
illumination changes, local signal disturbance, visual sensors. It did
enabled also to shed the light on critical parts that should be
investigated for improving the current version.
Work in
collaboration with Sinisa
Segvic, Albert
Diosi, Patrick
Gros and François
Chaumette.
Landmarks management and tracking during navigation
Developping an autonomous vision-based navigation system imposes to deal with the recognition of some visual information than can be extracted from the views acquired by the camera. These landmarks have also to be put in relation with the ones detected onto the image path. The use of a differential tracker manages to estimate within each acquired view the position of landmarks (Harris points) that were visible in the previous image. We are currently working on the improvement of such trackers to suit well to mobile vehicle characteristics (some results here).
During the navigation, the environment observed by the camera
continuously change. Therefor, the problem of automatic update of
visual landmarks must be considered. A method based on image transfer
is developed for detecting the apparition of landmarks from the image
path within the camera field of view. Within this framework, no 3D
model is needed, since a local reconstruction is sufficient. This
approach has been tested and validated during the real experiments
performed with the Cycab.
Work in collaboration with François
Chaumette and Sinisa
Segvic .
Qualitative visual servoing
An approach derived from classical visual servoing, called qualitative visual servoing, is proposed to control the motion of a robotic system. The main property of this control law is that it does not require visual features to converge exactly towards a desired value, but rather towards a confident interval. We are currently studying the theoretical properties of this original scheme, and through this work, considering the behavior of control laws that use a weigthing matrix to activate or inactivate visual features.
As an introduction, the qualitative visual servoing has been
presented and firstly used for controlling the visibility of points
during a positioning task. Further works engaged are investigating the
other applications that could benefit from this concept.
Work in
collaboration with Nicolas
Mansard and François
Chaumette.