There are many different forms of human-robot interactions, allowing a team of humans and robots to take advantage of skills of each team member. A developing area of research is focused upon the potential of robot swarms working in emergency settings. In particular we consider a swarm of robots that are capable of supporting and enhancing fire fighting operations co-operatively. This paper outlines some of the key characteristics of this emergency setting, robot swarms within it and the work conducted to develop effective human-robot interaction.

EXOSTATION is a project aiming at building a complete haptic control station, which allows the operator wearing an exoskeleton-based haptic interface for the human to remotely control a virtual slave robot. The video presents the results achieved with the final system. It is composed of four main components. A human-arm exoskeleton used as a master haptic interface, its controller, a simulated virtual slave robot and a visualization tool. Demonstration scenarios are presented in the video and show the capabilities of the system.

EXOSTATION is a project aiming at building a complete haptic control station, which allows the operator wearing an exoskeleton-based haptic interface for the human arm to remotely control a virtual slave robot. This paper briefly describes the various components : the Sensoric Arm Master (SAM), a portable haptic exoskeleton, the Exoskeleton COntroller (ECO), the slave simulator, simulating an anthropomorphic manipulator and a 3D visualisation client. Several teleoperation control strategies (impedance, hybrid control, 3-channel) have been tested and compared in order to evaluate their performances. The last has shown the best behavior in term of haptic feedback. Finally, a focus is made on the application, and how various manipulation and operation tasks can be performed to assess the system's performances (contact wall, objects manipulation, screwing). Users who tested the system were very impressed by the easiness of operation with the exoskeleton and felt the advantages of a force feedback information.

In this paper we propose an on-line incremental vision-based topological map learning for AIBO robots. The topological map is represented through a graph, where the vertices encode views from the robot's environment and the edges the spatial relationship between these views. The views are represented through their SIFT keypoints. The proposed map learning method has been successfully applied to a homing application.