Semantic Representation for Communication Between Human and Wireless Robot

We have developed a human-navigation-inspired and semantic map-building for communication between human and wireless robots. Four human navigation strategies (dead reckoning, egocentric place recognition, and reorientation) are integrated in our proposed method as an egocentric-semantic topology representation, egocentric measurement representation, and reorientation rules, respectively. In experiments, we successfully created a semantic representation and estimated the location of the robot by changing inaccurate distance and angle to approximate measurements using our proposed computing method. Semantic representation building took place in a 14 × 26.5 m corridor. The proposed method enables a robot to navigate effectively in a service environment using navigation schemes inspired by humans and their actions.