Abstract: Objective It aims to meet the demands of digital monitoring and motion simulation of robots in the additive manufacturing field. Methods By combining theoretical and experimental methods, taking KUKA KR6 industrial robot as the research object, this paper explores the application of digital twin and kinematic simulation in additive manufacturing robots. Results A four-dimensional architecture of the additive robot digital twin system covering physical entity layer, network layer, data core layer and human-machine interaction layer is proposed. A virtual-real mapping environment for additive manufacturing is constructed from both geometric and data dimensions, achieving real-time interaction and synchronization between physical entities and virtual models. In terms of kinematic modeling and trajectory planning, this paper has completed D-H parameter modeling of the robot and conducted forward and inverse kinematic analyses, verifying accuracy of the modeling. Meanwhile, this paper integrates functions such as data fusion, collision detection, historical data traceability, and online file transmission. Based on the improved hybrid hierarchical bounding box collision detection algorithm, accuracy and efficiency of collision detection are enhanced. Through online data transmission and historical retrospection, multi-source processing of data has been achieved. Conclusion The system proposed in this paper can achieve real-time monitoring of the additive process and kinematic simulation of the robot, effectively improving safety and virtual-real interaction ability of the additive process.