Aiming at the problem that the unmanned surface vehicle (USV) needs to recognize the ship’s interference intention in the process of obstacle avoidance, a ship interference intention recognition model based on Gaussian mixture model (GMM) and continuous hidden Markov model (CHMM) was proposed. Firstly, considering the complexity of intention recognition, based on the speed obstacle model of USV and ship, the motion characteristic parameters such as multi period interference coefficient were proposed as the input of the model; and considering the continuity of ship motion, GMM was used as the observationstate transition probability distribution. Then, the training data were trained by BaumWelch algorithm to obtain the intention recognition model. Finally, the prediction samples are obtained and the accuracy of the model is evaluated. The results show that the model has good recognition effect and can improve the safety of USV.