Abstract: A human-robot interaction system is designed with the hybrid gaze brain-machine interface and the shared control strategy, to make the user continuously control the robot end-effector in 2D space with his/her gaze and thought, meanwhile obtaining timely assistance from the machine intelligence in the task of avoiding obstacles and reaching target objects. Firstly, the movement speed of the robot end-effector is adjusted continuously and proportionally by the motor imagery strength of the user, in order to increase the user's sense of control and his/her engagement in the task. Next, a shared control strategy is proposed for controlling the movement direction of the robot end-effector, which dynamically fuses the direction commands from the user obtained by gaze tracking and from the robotic system for avoiding obstacles and reaching target objects. Such a shared control strategy adaptively adjusts the assist level for the user, so as to reduce the mental workload of the user and improve the success rate of completing the task. Finally, with respect to the constructed human-robot interaction system based on the hybrid gaze brain-machine interface and the shared control, experiments are conducted to verify its effectiveness.