Hidden Markov model-based occupancy grid maps of dynamic environments

摘要: For a reliable localization in dynamic environments, robust representation is of vital importance to obtain an accurate position estimation. This paper introduces hidden Markov model-based approach representations environments. The model uses occupancy grid maps created at different times as observations. involves map registration process for pre-processing align new observations impacting the representation. It combined feature-based method and NDT-based refinement. representation, used estimate probabilities static states cells based on updated with each observation using iterative propagation algorithm. Experiments real world radar data demonstrate that algorithm this provides more performance than standard approach.