Error modeling of Laser Range Finder for robotic application using time domain technique

Present day mobile robots are meant for very precise applications. For very precise applications of mobile robots, accurate estimation of inertial parameters depends upon the accuracy of mathematical model & as well as accuracy (error characteristics) of the individual sensor measurements. Sensor measurements are prone to various errors, which necessitates the detail modeling of sensors for estimation of useful signals from the noisy sensor measurements. Detail error modeling is essential to understand, identify & characterize the different types of noises present in the measured data using available mathematical techniques. This paper illustrates the frequency and time domain analysis techniques for characterization and identification of various noises present in the Laser Range Finder (Model: LMS200, SICK, Germany) measurements and their contribution to the overall noise statistics. A detailed methodology based on stochastic discrete time model is presented for Laser Range Finder error modeling.