Abstract: We carry out the research on teleoperated humanoid servo robots for its suitability in replacing human operators in dangerous environments. Firstly, based on the biological characteristics of the human upper limb, we design a slaved humanoid servo robot, and analyze its theoretical motion space using the Monte Carlo method. Secondly, based on the kinematic characteristics of the human upper limb, we construct a wearable human data acquisition system in combination with micro-inertial sensors and flexible curvature sensors to enable the system to obtain human motion information. Thirdly, we use the piecewise linear interpolation method to optimize the master-slave attitude-motion mapping capability, and analyze the system performance based on the time sensitivity and angular delay. Finally, we conduct experiments in which the humanoid servo robot model machine followed and carried objects. The results show that the proposed model machine can satisfactorily meet the precision and robustness requirements of the humanoid servo. The humanoid servo robot has the same motion space as human in particular sectors and is able to replace humans in performing dangerous tasks.