Human-Robot Interaction: Papers from the AAAI Fall Symposium
Alan C. Schultz, Chair
While we talk about "autonomous" robots, we realize that for virtually all of the future applications for autonomous vehicles, humans will always be in the loop at some level. The level of interaction can vary -- for example, a tactical mobile robot might have regular and frequent interactions with a human, while a planetary exploration robot might take supervisory instructions only occasionally -- yet the human is still part of the system. Even more, many envision a tighter collaboration between future intelligent vehicles and humans where the interactions are more mixed initiative, with all parties contributing according to their expertise.
In much current research, we have concentrated on robotic architectures and then put user interfaces on as means of monitoring and controlling the robotic platform. To develop a more synergistic system it is necessary to develop robotic architectures that accommodate the human in the loop from the beginning. However, this architecture needs to be designed to support dynamic autonomy. It should support human intervention when needed as well as supporting more autonomous behavior when the robot is capable of such actions.