Hybrid Mapping for Mobile Robots Based on Hierarchical Mixed Feature

To provide more inviroment information, this paper presented a hierarchical mixed featurebased hybrid mapping method. According to semi-structured environment, geometric feature was divided into regular and non-regular areas, and an improved GNN algorithm was introduced to accomplish nonregular area clustering simultaneously. Furthermore, Bayes algorithm was used for dynamic information processing. Experimental result shows that this method is effective and practical.