Environmental Modeling and Identification Based on Changes in Sensory Information

Adaptability to various environments is needed for a robot that supports our lives. Environmental identification is important for a mobile robot that works in multiple environments (e.g., different rooms). We present an environmental modeling method based on state representation, which represents a change in sensory information. Our model enables the mobile robot to identify which environment it is in. The results of experiments on a real mobile robot with only low-sensitivity infrared sensors show the effectiveness of our method, and a comparison between our method and a conventional one shows that ours has higher performance.