Abstract:

In vocational and technical training, remote instruction involves a trainer guiding a learner through a hands-on procedure from another location. Video conferencing is often used for this purpose, but it provides limited support for linking the trainer's verbal guidance to the relevant equipment component, workspace location, or procedural step. The guidance situation becomes more demanding when explanations are given in one language while technical terms and component labels appear in another. The learner must then connect the instruction, the term, and the component at the same point in the task. This added coordination can make the procedure harder to follow and increase the mental effort required. Human-computer interaction research has examined remote guidance for physical tasks, yet bilingual vocational instruction remains underexplored. This paper presents a cloud-based, spatially anchored augmented reality framework built around trainer-cloud-learner architecture. It links bilingual guidance to the relevant physical component and active procedural step, while the trainer controls progression and the cloud coordinates the shared procedural state and delivers the corresponding guidance. By keeping the instruction, terminology, and point of action aligned within the task view, the framework is intended to support task fluency by reducing avoidable sources of mental effort.