Welcome in Philippe Martinet home page

Research

During my carrier, I have setup 4 research teams in collaboration with my colleagues:

CHORALE (Inria: 2018-...): Collaborative and HeterOgeneous Robots interActing in Live Environment
ARMEN (LS2N: 2017-...): Autonomie des Robots et Maitrise des interactions avec l'ENvironnemt
ROSACE (LASMEA: 2006-2011): RObotique et Systemes Autonomes ComplExes
RCS (LASMEA: 1999-2006): Robots and Complex Systems

My main research activities concerns Modeling, Control and Perception:

Task Based Modeling and representation of the environment:
Robotic tasks can be defined by controling the interaction with the environment while controling the internal dynamics. When using multiple information and/or multiple sensors, a set of parallel interactions are created, which can be modelized by a virtual parallel robot called Hidden robot. By studying the proprerties of this hidden robot, we are able to analayze and study the control laws. We strongly believed that environment must be defined knowing the task to be done.
Mobile and Autonomous Robots :
Intelligent and connected autonomous vehicles (or mobile robots) are the next revolution technology which will be part of the human environment. Robots must be not only automated, they must be also autonomous and safe in order to face and handle difficult situations. Evolving among humans requires to be not only safely, but also to be proactive in order to accomplished transportation tasks. Multi robots architecture allow to enhance perception and action for robotic systems.
Visual Servoing and Sensor Based Control of Complex Systems:
Ominidirectional visual servoing using different features (lines, points, ...) can be used for large environment. Complex systems like parallel robots, humanoïd robots, multi arms robots, multi robots systems, UAVS, and combination of them are main targeted robots. Using adaptive control, predictive control and online learning, allows to anticipate and adapt the system to the evolution of the dynamic environment.
Sensor Modeling :
When sensing the environment, it is often required to use large field of view cameras, omnidirectional camera, even 360degres camera. Generic camera models can be proposed in order to preserve the main geometrical properties of the data. For instance, the unit shperical model has been demonstrated to be candidate for the fisheye camera model, and an enhanced version has been developed. New sensors technology requires always to setup new model, and propose generic model. We all dream for a generic multimodal model able to be used for different sets of sensors.
Robotics & AI :
Artificial intelligence is present at the time in the way of defining and designing the tasks, in the way of representing the environment using multi-layers (metric, topological,semantic, social, ...), in the description of the knowledge associated with the task and its evolution through experience (learning by reenforcement, or deep learning), and in the cognitive mechanisms put in place to manage the short, medium and long term information. It is also present in the algorithms and exploration strategies, in algorithms collaborative perception and command for multi-robot systems, ...

Chronology

November 2017: I moved to Inria Sophia Antipolis in the Lagadic Project for a 5 years period. I am still research collaborator in the ARMEN team. since April 2018, I am the leader of the new team called CHORALE (Collaborative and HeterOgeneous Robots interActing in Live Environment).
January 2017: In collaboration with my colleague Sebastien Briot, we have created the ARMEN team. I was co-leader of the team. The main research axis were:
- Robots design and environments
- Robot Interactions with the environment
- Robot perception of the environment
September 2011:I have joined the robotic team at IRCCyN in 2011. My main research interests were :
- MICMAC : Modeling, Identification and Control of Complex MAchines
- VISIR : VIsual ServoIng of Robots
- Intelligent Vehicles : Modelling, Control, Perception, Integrity, Safety
- ARMS : Modelling, Control, Perception for multi-arms manipulation (humanoid, dual arms, multi arms platform)
2006-2011: I was the leader of the team ROSACE (I created it) in LASMEA-GRAVIR. In LASMEA-GRAVIR, the team "ROSACE" was concerned by many research activities which are developed or haved been developed in the past :
- VISIR : VIsual ServoIng of Robot Visual Servoing (Position, Image and Hybrid Based, Omnidirectional), Multi-Sensor Based Control, Force-Vision Coupling, Robots (Manipulator, Mobile, Aerial)
- AGV : Autonoumous Guided Vehicle Control (Non Linear, Adaptive, Predictive, Robust), Enhanced Mobility (Sliding and Slipping), Uncertain Dynamics, Monitoring, Robots (Mobile, AGV, All Terrain vehicles, Agricultural Vehicles, Platoon, Multi-Robot System)
- MICMAC : Modeling, Identification and Control of complex MAChines Kinematic Identification, Dynamic Identification, High Dynamic Modeling, Dynamic Control, Vision based Control of Parallel Robot, Robots (Parallel Robot, High Speed Machine Tools)
- RCA : Redundancy, Control and Autonomy Humanoid robot control, Multi Arms control, Redundancy and polymorphism, Robots (Humanoid robots, Service Robots, Polymorph Robots)
2006-2007: I was visiting professor in Intelligent Systems Research Center in Suwon, Korea.
1999-2006: I was the leader of the team RCS (I created it) in LASMEA-GRAVIR. My main research interests were :
- Sensor based control of Robots
- Autonomatic Guided Vehicles
- Modeling, Identification and Control of Complex MAchines
- Humanoid and Multi arms robot

I am also interested in other research activities which are currently developed or haved been developed in the past :

AVSI : Active Vision and Sensor Integration
PAVA : Parallel Architecture for Vision Applications

AVSI : Active Vision and Sensor Integration

Current research activity (since 2000)

Current actions concern :

History of research activity in LASMEA

Some past actions have been done by Jozeph Alizon (CR CNRS/LASMEA).

PAVA : Parallel Architecture for Vision Applications

Current research activity (since 2000)

No current action.

History of research activity in LASMEA

Past actions concern :

Others Parallel Architectures for Vision Applications have been developed in the LASMEA : PRIVE1, PRIVE2, TRANSVISION (T800, T9000, Dec Alpha), Bewulf Ossian Machine (G4), ...
For more details or advanced informations, please contact Jean-Pierre Derutin at LASMEA.