The present disclosure is applicable to robot technology. A method for robot fall prediction, and a robot are provided. The method includes: searching a weighted value of a center of gravity of the robot corresponding to a posture of the robot, according to a preset first corresponding relationship; correcting an offset of the center of gravity of the robot based on the weighted value of the center of gravity of the robot; correcting an acceleration of the robot based on an offset direction of the center of gravity of the robot; and determining whether the robot will fall based on the corrected offset of the center of gravity, the offset direction of the center of gravity, and the corrected acceleration of the robot. The present disclosure improves the real-time performance and accuracy of the prediction for the fall of a robot through the fusion calculation of various data.