Aiming at the problem of low traffic efficiency of unmanned vehicles caused by complex characteristics in off-road environment, this paper combines the knowledge of human-like driving vehicles, designs a trajectory planning method based on human driving prior knowledge, combines environmental perception data, realizes efficient path planning of unstructured roads in off-road environment, and compares it with the method based on LSTM. The results show that the proposed method is more accurate than the LSTM-based path planning method.