Robot Path Estimation from Odometry Measurements: Loop Closure Based on Autocorrelation

The problem of estimating the Robot path when covering closed trajectories using only the wheels odometry information is met in indoor low-cost robotic applications. Random and systematic errors related to the measured angle accuracy, deviations from the desired path and wheel radius in robot turns, introduced significant errors in the process of estimating the actual path. In this paper the autocorrelation function is used for accurate estimation of the path periodicity and to reconstruct the robot path. Real-life experiments using a real robot showed minor differences between the estimated and the predicted path.