Mead

We present an overview of the control, recognition, decision-making, and learning techniques utilized by the Interaction Lab (robotics.usc.edu/interaction) at the University of Southern California (USC) to enable autonomy in sociable and socially assistive robots. These techniques are implemented with two software libraries: 1) the Social Behavior Library (SBL) provides autonomous social behavior controllers; and 2) the Social Interaction Manager (SIM) provides probabilistic models to recognize, reason over, and learn about human behavior. Both libraries are implemented in the Robot Operating System (ROS; www.ros.org)