Disambiguating Robot Positioning Using Laser and Geomagnetic Signatures

This article proposes a technique for position disambiguating of mobile robots using laser and geomagnetic signatures in indoor terrains. Referring to the geomagnetic signature, magnetic field disturbances are used as unique recognition signatures. First, we gather information about the magnetic heading and the laser mapping as our robot traverses a path between start and goal points of the workplace using the vector mapping paradigm (VMP) [1]. Second, we employ a spectral analysis, based on the discrete cosine transform (DCT), to code and to compress geomagnetic and laser signatures, and then store this information. Finally, as the robot traverses a vector, it gathers the information from the compass and the laser scanner, converts it into DCT form and matches it with the pre-stored data. If a match is found, the robot can deduce its current position. The results show how the robot can distinguish accurately its position using these signatures.