One major application field of robotics in the future is service robotics where service tasks are performed autonomously or semi-autonomously by robots. While numerous domains can be listed in the industrial context (i.e. in an environment which has been designed or adapted for robot operation and where only professionals respecting very strict rules are allowed), there is a large number of potential applications where the robot should perform servicing for the public or in the presence of the public. There are also very interesting and very promising applications where the interaction can even be closer and where the robot assists (physically) a human.

This is precisely what is of interest here. We want to focus on all the topics linked to situations where the robot performs its tasks in interaction with humans. The human can be an operator, a user of a service performed by the robot or simply a user (a passer-by) of the same environment.

While substantial advances have been made in robotics resulting in a technology which has already great potentials for real applications, there is still a lot of work to be done to effectively open applications where a robot and a human can co-exist.

Let us give a simple example: when a cleaning robot encounters people during its cleaning task, how should it react and behave? What kind of interaction should it have with these people? How can it distinguish a human from an obstacle? How can it inform of its actions and goals ? how can it ask people to move aside ? How can it understand the situation and the goals of the humans around it ...?

More generally, we would like to tackle problems such as navigation in presence of humans, transporting humans, handing out objects to humans, manipulating and transporting loads in cooperation with humans, assisting handicapped or elderly humans...

Addressing such problems is not an engineering work which involves nice or ad hoc interfaces and more robust and more reliable implementation of existing state-of-the-art results. It is a real scientific and multi-disciplinary challenge which covers numerous aspects: mechanical design, sensors, control, perception, decision, communication interfaces.

We should build ``soft robots'' which should ``comply'' and never harm human. We should devise robots able to effectively deal with the variety and dynamics of environment where the public is present. Depending on the applications, such robots should interpret human gestures and movement or even human plans and goals which may interact with the robot goals. In other situations, they should filter and understand the user requests depending on the context. The robots should also be able, when necessary, to inform about their actions and goals and to get relevant information from people.