As a key enabling technology to advance human–computer interaction, human motion digital twin focuses on the behavior of human ontology in human–computer interaction. The combination of robot capabilities with human intelligence is gradually showing promising potentials. Teleoperation interaction is a mode of direct sharing of operation intention in human–robot interaction. However, due to the differences between the operator and the robot in movement, motion space and joint degrees of freedom, the teleoperation system cannot share the intention directly, accurate expression of human operation intention puts forward higher requirements for hardware resources and algorithm strategy. This work designs a generalized IoT-enabled telerobotic architecture by integrating human-motion-capture subsystem with robot-control subsystem. This work solves the motion trajectory of the end joint of the human body, according to the difference of human–robot motion characteristics and motion space, the motion-motion mapping strategy is designed to clarify the motion path planning of the teleoperation robot, which provides support for describing the human motion digital twin. As specific examples of application, the proposed architecture is applied to complete tasks based on dual-arm motion capture and mapping for telerobotics in laboratory and intercontinental scene.