Enhancements in a Social Robotic Simulator for Indoor Environments

Social robotics represents a branch of the Artificial Intelligence (AI) area, which aims to develop robots to work in unstructured environments in partnership with human beings. Research in AI and particularly in cognitive science are catalysts in the development of social robotics as they model the ability to obtain, process, store information and direct actions performed by a robot. In this article, a simulator, Robot House Simulator (RHS), is proposed to validate robotic systems controlled by cognitive architectures. The main objective of RHS is to minimize the effort made by the developer of cognitive systems in obtaining information associated with the actuators and sensors of the robot. For this, the simulator makes use of high-level abstraction commands and the standardization of sensory information through the OntPercept ontology, which is being also developed. Some experiments that demonstrate the simulator's ability to provide a validation environment for cognitive architectures directed to social robotics are presented. This environment is a pioneer in the integration of a simulator and a cognitive architecture, as well as providing rich sensory information, highlighting the unprecedented modeling available for the senses of smell and taste.