The invention discloses a robot trajectory tracking method based on the Skinner conditioned reflex automaton, relates to the field of mobile robot trajectory tracking, and in particular relates to a robot trajectory tracking method based on the Skinner conditioned reflex automaton. The present invention first establishes the operation and state set of the robot, and establishes the corresponding probability set from the state to the operation, and makes it conform to a uniform distribution; then, randomly selects an operation, calculates the corresponding position change, and then calculates according to the distance from the target track. The orientation function adjusts the action probability distribution according to the orientation function value according to the operant conditioning theory, and calculates the system entropy; when the system entropy tends to the minimum value, the learning ends. At this time, the probability matrix is optimal. The invention can well simulate the operation conditioned reflex behavior of humans and animals, and improve the intelligence level of the robot, because it has strong self-learning, self-organizing, self-adapting capabilities, independent condition parameters, and successfully tracks tracks.