The work aims to develop an interaction gesture design method that aligns with user cognition, so as to enhance the user experience of fully autonomous vehicles (FAVs) navigation without steering wheel control. Initially, the navigation gestures for both internal and external scenarios of FAVs were collected by a user generated design test. Then a navigation gesture collection applicable for FAVs was constructed based on consistency comparison. Finally, the usability of the navigation gesture collection was evaluated through card sorting and simulation testing. The results from card sorting and simulation testing indicated that the gesture design method which combined user participatory design and consistency calculation achieved a positive user experience expectation for FAVs. It also affirmed the feasibility of navigation gestures for both internal and external scenarios of FAVs. This research validates the effectiveness of gesture interaction in enhancing the user experience in internal and external scenarios of FAVs, and provides a new direction for the user experience research of FAVs.