In order to prevent and reduce the number of accidents occurred in rainy nights ,this study focuses on how the rainfall intensity impacts on driver's visual cognition distance so driving speed ,stopping sight distance and adviso‐ry speed can be determined .38 drivers were recruited and the actual test‐driving (71 tests) was conducted in rainy nights with different rainfalls .All tests were grouped into 25 groups according to different speed and rainfall intensity using the cluster analysis method .Driver's visual recognition distances of the black obstacles at different speeds were collected .The study developed a model of visual distance recognition distance with variables of rainfall intensity and driving speed using the method of curved surface regression .The results show that the correlation coefficient is 0 .983 2 and effective degree coefficient is 0 .94 ,which confirms the strong relationship between visual cognition distance and the instantaneous rainfall intensity and speed .The results also indicate that the visibility distance decreases when the rainfall increases ,and visibili‐ty distance decreases when the driving speed increases .Under an extreme condition when drivers drive very fast in a heavy rain night ,the visibility distance will decrease greatly and allowed reaction time will be shortened rapidly ,which may cause an accident .