SAMPEN Associated team project Homepage

Self adaptive mobile perception and navigation (2014-2016)

Sampen is an Inria associated team project accepted in 2014 for 2 years. It is funded by INRIA. The coordinator of the project is Anne Spalanzani from UPMF University.

The aim of the project is to propose a self-adaptive system of perception combined with a system of autonomous navigation. Usually, systems of perception rely on a set of specific sensors and a calibration is done in a specific environment. We propose to develop some methods to make perception systems adaptive to the environmental context and to the set of sensors used. This perception, that can be embedded on the mobile robot as well as on home structures (wall, ceiling, floor), will be helpful to localize agents (people, robot) present in the scene. Moreover, it will give information to better understand social scenes. All information will be used by the navigation system to move with a behavior that fit the context. The partners for this project will put in common their expertise. ICeiRA will provide their technologies on sensors, force motor command, and its ability to build robotic systems [Luodec11] [Luosep11] [Luooct11] [Luodec08] [Luodec08b]. Emotion-INRIA will bring their expertise in perception of dynamic environments and social conventions, and in planning and navigation in dynamic and human populated environments [perro12] [Rios11] [Vasq09a] [Vasq09b][Fulg09]. Lagadic-INRIA will bring their expertise in visual servoing of mobile robots [Cher13a] [Past13] [Cher13b] as well as in visual tracking [Sekk13] [Comp06]. Phoenix-INRIA will provide their expertise in calibration of complex systems and smart environment [Daney05][Gayral13].

Research directions

This project focuses on two main directions :
1.self-adaptive system of perception.
2.autonomous and semi-autonomous navigation in a human environment

Anticipated challenges

Concerning the self-adaptive system of perception, the main challenges are:
- Fusion of heterogenous sensors
- Self calibration and choice of reference constraints
Concerning the autonomous and semi-autonomous navigation system, the main challenges are:
- Localization using heterogenous sensors
- Navigation following social conventions

intended approaches, methodologies

The self-calibration of heterogeneous sensors is communally based on the minimization of an error obtained using a shared redundant observation of a reference object or a well-identified environment. The proposition is to study the type of observation (geometrically and cinematically) to deduce a criterion on the numerical quality of the identification of sensors positioning parameters. The results should also serve to design the placement of the sensors as a function of a dedicated task because, in this case, at the opposite way, the information obtained by sensor fusion have to be the less sensitive as possible to sensors positioning parameters.
The localization issues will be mainly addressed through robotic vision systems with the help of 2D/3D cameras By using localization, navigation relies on a data fusion process that also exploits data obtained from proximity and odometric sensors. The redundancy of information obtained from a set of heterogeneous sensors will guarantee the dependability required by such autonomous systems. In addition, the resulting trajectory of the autonomous wheelchair would have to take into account the dynamic environment inherent to the targeted human environment. In particular, the planning process will include cognitive elements to elaborate motions that remain socially permissible.
The Navigation will take into account social conventions adapted to Taiwan and/or to France. Navigation will be based on the perception embedded on the robot but also on the sensor network. Some socially acceptable behavior will be developed on the robot so that it can be better accepted by humans.

Last modified the 4th of november 2014
Anne Spalanzani. INRIA Rhône-Alpes, 655 Avenue de l'Europe, 38334 Saint Ismier Cedex, France Phone: +33 4 7661 5292 -- Fax: +33 4 7661 5477 Email: Anne.Spalanzani"followed by"@"and by"inria.fr