ABSTRACT
This chapter introduces different instances of shared autonomy study in social robotics. It describes a rule-based semi-autonomous system, and show how such semi-autonomy could contribute to improving the performance of robotic systems. The chapter explores a technique to divide operators’ attention among multiple robots. The robot is developed with a “network robot system” approach. That is, a humanoid robot is connected with sensors embedded in the environment over network. The Semi-Autonomous Robot System is designed with the idea of having three layers: reflective, behavioral, and reactive. This design is informed by D. Norman’s human model. For the system is used Robovie, an interactive humanoid robot characterized by its human-like physical expressions and its various sensors. The conceptual purpose of the reactive layer is to achieve safe interaction and lifelike behavior. In order to provide safe interaction, the robot’s locomotion and motors stop when an object contacts the robot’s bumper or an overload of any motor is detected.
