Abstract: A gait planning method for the lower-limb rehabilitation training robot with the physiotherapist interaction is proposed to solve the problems of individual differences and pathological conditions for stroke patients. In the complex environment where the physiotherapist, the body weight support treadmill training robot and the patient coexist, the physiotherapist wears on the master exoskeleton mechanism and walks directly to plan the space-time gait parameters. It integrates rehabilitation experience of the physiotherapist and the assessment of patients. Firstly, the kinematics model is established based on the screw theory, and the motion mapping is realized from the physiotherapist space to the robot space. Then, the joint moving trajectories of the robot, the gravity adjustment trajectory of the weight support mechanism and the walking velocity of the treadmill are planned as a whole. Finally, the effectiveness of the space-time gait planning method is verified by real-time acquisition of the physiotherapist gait parameters, motion mapping test and robot trajectory tracking tests. Results show that the planning angles of the hip and knee joints are in the range of human joint motion, and the angular velocities of the joints are smooth. The joint trajectory planning and gravity adjustment planning conform to the physiological characteristics of human walking. The participation of the physiotherapist achieves personalized gait planning in the progressive rehabilitation training.