Route natural language processing method for robot navigation

Abstract

With the development of artificial intelligence, robot navigation based on natural language processing is concerned by more researchs. At the present stage, there are few studies on route natural language processing (RNLP), and there is no professional corpus, so the method of route natural language processing is generally based on a small amount of restricted corpus. This paper puts forward a method of semantic extraction based on semantic role labeling. Firstly, the simulation environment is built, and the route natural language corpus is built. Then, based on the analysis of the route natural language corpus, this paper proposes several common chunks of route natural language, and carries out the chunking experiment. Finally, semantic role labeling (SRL) is performed on the basis of the chunking. According to the results of semantic role labeling, navigation intention map is generated. The experimental results show that this method is suitable for the semantic extraction of route natural languages with non restricted.