A system and a method for controlling the robot. The method includes determining, based on feedback data received from the robot, first position information of a joint of the robot. The feedback data indicates a movement of the joint. The method also includes determining second position information of the joint based on sensing data received from a sensor. The sensing data indicates a relative movement between the joint and a second object to be aligned with the first object. The method also predicts a target position for aligning the first object with the second object.