In order to solve the problem of tracking control of the target trajectory by the exoskeleton robot, the trajectory tracking control of the lower extremity exoskeleton robot for rehabilitation training is studied. According to the characteristics of the exoskeleton model, the Lagrange method is used to establish the swing phase dynamics model, and the swing phase of the one-leg support gait is analyzed. The control model is established by using Simulink, combined with the fuzzy PID algorithm and the traditional PID algorithm, the response, square wave signal, and sinusoidal signal are compared and simulated, and the simulation results are evaluated. The simulation results show that compared with the traditional PID algorithm, the fuzzy adaptive PID has the advantages of small overshoot and strong stability.