Abstract
A method for controlling robot motion includes: determining a remaining duration of a current stepping cycle, which is a duration from a current moment to an end moment of the current stepping cycle; predicting, based on the remaining duration, an expected velocity, a first center of mass position and a first instantaneous velocity of the robot at the current moment, and a state space equation, a second center of mass position and a second instantaneous velocity of the robot at the end moment; determining, based on the expected velocity, the second center of mass position and the second instantaneous velocity, a stable foothold position corresponding to the current stepping cycle; determining, based on the stable foothold position and the second center of mass position, at least one motion parameter corresponding to at least one joint in the robot associated with stepping; and controlling the robot based on the at least one motion parameter.