Simultaneous learning of motion and sensor model parameters for mobile robots

Abstract

Motion and sensor models are crucial components in current algorithms for mobile robot localization and mapping. These models are typically provided and hand-tuned by a human operator and are often derived from intensive and careful calibration experiments and the operator's knowledge and experience with the robot and its operating environment. In this paper, we demonstrate how the parameters of both the motion and sensor models can be automatically estimated during normal robot operations via machine learning methods thereby eliminating the necessity of manually tuning these models through a laborious calibration process. Results from real-world robotic experiments are presented that show the effectiveness of the estimation approach.