Intelligent obstacle recognition is the essential technology of the self-driving car

which has been a popular research so far. Based on the basic theory and method of Extenics

a research is carried out by using the extension recognition method

to solve contradictory problems. The general procedure of extension recognition method is provided

combining qualitative analysis with quantitative calculation. This method is applied on obstacle recognition of the self-driving car and a strategy obtained that conform to the recognition requirements. The case analysis proves the feasibility of this method. In the meantime

this method offers a research idea for the development of extension recognition system and has a broad application prospect.