The autonomous social robot control based on the situation analysis

Abstract

Abstract The problem of the autonomous task solution by the robotic system is considered in this paper. An approach based on the situational assessment and analysis is proposed for this. This approach makes it possible to plan the robot actions from the point of minimizing the situations informational uncertainty. The hirachical semantic descriptions of the target task, environment and the robot’s state were proposed. Within this approach the situational uncertainty in terms of informational entropy estimates. Each action, including planning, should be aimed to reduce entropy. An example of the proposed approach implementation for the object dellivery task was demonstrated. The efficiency ac the approach was shawn.