HiMAR: a hierarchical context-aware mobile AR assembly for large-scale cable products

Existing augmented reality (AR) cable assembly mainly provides step-by-step intuitive instructions within a specific workbench, and it is hard to establish a holistic AR guidance from global mission planning to local instruction for large-scale assembly task. To bridge this gap, this paper proposes HiMAR, A hierarchical context-aware mobile AR assembly for large-scale cable products. Firstly, a self-contained multi-sensor-based motion tracking method for a resource-constraint mobile AR platform is presented, enabling a real-time alignment between the dynamic virtual instructions against the actual workplace. Then, a cognitive AR mission planning method against the ongoing assembly procedures is proposed, as well as the spatial–temporal data-driven verification for successive assembly procedure automatically. In addition, a context-aware mobile AR assembly with cable character recognition is achieved, followed by hierarchical AR guidance from global path navigation to local intuitive visual instruction for large-scale cable assembly tasks. The novelty of this work is focused on the deployment of hierarchical context-aware mobile AR instructions using the same device that is being applied to deploy on large-scale assembly tasks involving lots of workstations. Finally, experimental results illustrate that the proposed HiMAR can achieve superior task performance and alleviate the cognitive workload, resulting in a more human-centric AR assembly.