Indoor mobile robot self-localization based on a low-cost light system with a novel emitter arrangement

Abstract This paper proposes a low-cost infrared (IR) LED system with a novel arrangement that enables the self-localization of an indoor mobile robot. The proposed system uses only low-cost IR LED emitters installed on the ceiling of a building and low-cost IR receivers, each equipped with a photodiode, located on top of the mobile robot for detection. The IR LEDs, which are driven by an on-off keying modulator with various frequencies, are used as active landmarks. The mobile robot localizes itself based on the IR LED signals it receives. Although it would be desirable to assign a unique ID to each emitter, this would be difficult to realize because of the limited number of available frequencies and the production cost resulting from the use of independent microcontrollers. Therefore, we designed a novel landmark configuration consisting of an IR LED array with unique ID-encoding capabilities based on a combination of different frequencies and the repeated use of each ID encoding LED to address the problem of limited frequencies. The novel landmark arrangement reduces the costs associated with production, ID encoding, and computation. A Monte-Carlo localization method with a belief function is utilized to estimate the position and orientation of the mobile robot based on the decoded ID information and received signal strength. The proposed system was tested in a real environment and the experimental results confirmed the validity and accuracy of the system and method.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.