Адаптация промышленной дополненной реальности для профессиональной ориентации с помощью контекстуальных и низкозатратных сценариев обучения

The structure of the modern economy is changing at a pace that significantly outstrips the inertia of educational institutions. New occupations appear, traditional ones merge or disappear, and the technological content of the work performed under a familiar job title may shift radically within a few years. Against this background, the instruments by which schoolchildren form their first ideas about the world of work remain largely unchanged: a lecture given by a careers counsellor, a glossy brochure produced by a sponsoring enterprise, or a guided tour conducted at a respectful distance from the actual production line behind a safety railing. None of these formats leaves more than a surface impression. There is no opportunity to try anything with one’s own hands, to feel the resistance of a tool, to experience the tempo of a real operation, or to encounter the small frictions that ultimately determine whether a profession suits a particular person. Without such hands-on action, any picture of a profession remains essentially bookish and is easily displaced by stereotypes drawn from films, social media, or family expectations.

The mismatch between the speed of industrial change and the conservatism of career-guidance practices generates measurable social costs. Students enter higher and vocational education with imagined images of a profession that bear little resemblance to the daily reality of the workplace, which contributes to high rates of dropout in the first two years of study and to subsequent lateral movement into unrelated fields. Employers, in turn, complain about the protracted adaptation of graduates and about the difficulty of forecasting the demand for specific competencies. Analyses of job-market postings for immersive-technology specialists likewise reveal a persistent gap between the skills advertised by employers and those formed by traditional schooling

A small but growing body of work has begun to explore how immersive technologies may contribute to this task. WebAR-based career-guidance quests deployed at the level of an entire school system have shown that even lightweight browser-driven experiences noticeably improve the engagement of adolescents with information about professions

What is new in this paper concerns three closely related elements. First, methods of visual action guidance that have already been thoroughly field-tested in manufacturing are for the first time systematically transferred to the domain of career guidance, with explicit attention to the change in pedagogical goals that this transfer requires. Second, the principle of minimal AR is shown to be viable not only as a technical curiosity but as the basis for large-scale events conducted on a deliberately modest budget. Third, a model of short AR scenarios designed specifically to trigger vocational self-determination is proposed, and its key parameters — duration, hardware, structure of the participant’s interaction — are described in a form that allows immediate replication at the level of a single school or an entire regional programme.

The aim of the work is straightforward: to construct a working model for transferring proven shop-floor AR solutions into the practice of school and college career guidance, while preserving their economic accessibility. The first objective is to compile and systematise the available data on the effectiveness of AR at industrial sites, with particular attention to those metrics — reduction of training time, decrease in error rates, relief of cognitive load — that have potential analogues in the educational setting. The second objective is to substantiate the visual-guidance method grounded in the idea of minimal authoring, distinguishing it both from conventional video instructions and from full-fledged industrial AR with three-dimensional tracking. The third objective is to formulate the principles by which industrial AR scenarios should be adapted for career-guidance tasks, recognising that the criterion of success here is not the flawless execution of an operation but the participant’s ability to form a substantiated subjective judgement about the profession.

2.1. Industrial experience

Augmented reality has long ceased to be a novelty in manufacturing and has become one of the standard instruments of industrial training and operational support

Comparable patterns are documented in many other domains, from the assembly of complex mechanical units

2.2. Effects

According to data published by Aidar Solutions and SkillMaker, the deployment of AR in industrial training reproduces a small set of consistent effects across very different projects. The most visible of these is the contraction of training time, which falls by up to seventy-five per cent for complex operations and, in the most striking case documented by Boeing SkillMaker, compresses what used to be a year of preparation into a single working week. Equally important, although less photogenic, is the reduction of errors: because the instruction is anchored to a specific physical component, it is much harder to confuse that component with a visually similar neighbour, and entire categories of substitution mistakes simply disappear. Finally, cognitive load decreases in a way that is felt rather than measured; a systematic review of controlled AR studies across a range of instructional settings confirms the effect and links it directly to improved task performance

It is worth emphasising that these three effects reinforce one another rather than acting independently. Faster training is partly a consequence of the lower cognitive load, since less mental effort is wasted on integrating fragmented information sources. Lower error rates feed back into faster training, because remedial work and the demoralising experience of repeated failure are reduced. The combined effect is a qualitatively different relationship between the novice and the task: the worker is allowed to be a beginner without being penalised for it, and the curve of competence becomes noticeably less steep. Similar reinforcement loops have been documented for the acquisition of psychomotor skills such as welding, where VR and AR systems progressively displace costly hands-on rehearsals without compromising the eventual quality of the technique

2.3. Limitations

The principal obstacle to a wider use of these technologies outside industry is cost. Full-blown AR systems with three-dimensional tracking are not only expensive in terms of the hardware itself but are also tightly coupled to highly specialised content. The cost of developing such content is high, the integration of the system into the existing curriculum is laborious and demands significant methodological work

3.1. The core idea

The principle of minimal authoring is simple to the point of appearing trivial. A person watches a reference video and immediately repeats with their own hands what is shown on the screen. Crucially, the video is not viewed in isolation on a separate device but is overlaid on the real working scene, so that the demonstration and the participant’s own action coincide in space and almost coincide in time. Unlike expensive industrial systems with three-dimensional tracking, the approach relies on what is sometimes called minimal AR authoring: content is produced quickly, cheaply, and without specialised equipment, since recording a clear demonstration of a task is something a domain expert can do with the same smartphone that the participants will later use to view it. A comparable design philosophy has been articulated in frameworks for motivational AR applications in vocational education, which deliberately favour lightweight content production over elaborate spatial modelling

Several practical features follow from this design choice. There is no requirement to model the workpiece or the tools in three dimensions, which removes the most expensive line item from any conventional AR project. There is no need to develop precise spatial tracking, because the participant aligns themselves to the demonstration by physical placement rather than by software registration. There is no obligation to install proprietary software on a controlled device, because ordinary mobile applications with camera access are sufficient. Each of these omissions individually saves resources; taken together, they shift the economic profile of AR from a capital-intensive industrial investment to something closer to the production of a short instructional video, while retaining most of the perceptual advantages of contextual overlay.

The model for transferring AR to career guidance can be described as a sequence of five stages that follow each other without sharp boundaries. The participant first observes the professional process through an AR overlay, which provides an initial structured impression. They then repeat the operations with their own hands, working alongside the demonstration rather than after it. The third stage consolidates this initial hands-on experience by allowing repetition without prompts, so that the participant can sense how much of the task they have actually internalised. The fourth stage explicitly grants time for reflection on what they have just felt, including aspects — boredom, satisfaction, physical fatigue, curiosity — that lectures and brochures cannot convey. Only after this sequence has been completed does the participant arrive at the fifth stage, where they are in a position to make an informed decision about whether the profession deserves further exploration. The logic is deliberately incremental: first one watches, then one does, then one thinks, and only then does one decide.

3.2. Advantages

Contextual AR is, in essence, a compromise, but a particularly well-balanced one. In terms of cost, it sits much closer to ordinary video instructions than to a professional industrial system, while in terms of pedagogical effect, it approaches that of full-fledged AR within the limits relevant to career guidance. The participant’s hands remain busy with the actual work, the reference is held in peripheral vision rather than competing for central attention, and there is no need for repeated focal switching between the instruction and the workpiece. The result is an experience that feels less like “watching a video about welding” and more like “welding while being quietly guided”, and it is precisely this difference in subjective register that does the pedagogical work.

An additional and often overlooked advantage of the minimal-authoring approach concerns the longevity of the content. Because scenarios are short, modular, and inexpensive to record, they can be revised whenever a procedure changes, a new specialisation appears, or a partner enterprise updates its equipment. A library of such scenarios behaves more like a living catalogue than a fixed curriculum, and its ability to keep pace with the surrounding economy is itself a pedagogical asset.

4.1. The specifics of career guidance

On the factory floor the goal of an AR-supported task is unambiguous: the worker must execute an operation correctly and without errors, and the system is judged by how reliably it supports that hard skill. In career guidance, the task is fundamentally different. What matters is not the precise reproduction of a movement but the formation of what may be called a subjective-evaluative attitude towards the work, that is, the participant’s own felt sense of whether the activity attracts or repels them, whether its rhythm matches their temperament, and whether the kind of attention it demands is one they can comfortably sustain. Precision becomes secondary, while the texture of experience becomes primary. Immersive-reality experiments aimed at informed career decision-making show that this shift in criterion has empirical substance: participants exposed to an embodied preview of an occupation report substantially more considered choices than those given conventional descriptive material

This shift in goal has methodological consequences that are easy to underestimate. An industrial AR scenario may legitimately be evaluated by the percentage of operations performed correctly on the first attempt; a career-guidance scenario must be evaluated by the quality of the participant’s subsequent reflection. The instrument is the same, but the criterion of success is different, and the design of the scenario must follow the criterion rather than the instrument. Evidence from immersive VR training for career adaptability points in the same direction: the measurable outcomes are psychological readiness and self-concept, not task execution

4.2. Adaptation principles

Three principles govern the adaptation of industrial scenarios to the career-guidance setting. The first concerns the simplification of essence: the participant should be helped to understand not so much the question of how exactly an operation is performed as the question of why it is performed in this particular way, since grasping the purpose of an operation contributes more to vocational self-understanding than reproducing it precisely. The second principle gives a clear preference to live video over animation, because a real production cycle captured on camera carries information about tempo, sound, posture, and material that no drawn diagram can convey, and these are exactly the cues by which a teenager intuitively decides whether a profession feels like their own. The third principle imposes a strict limit on duration: each scenario should last no more than five to ten minutes, because beyond that interval the attention of an adolescent participant predictably drifts, the experience loses its sharpness, and the fragile link between observation and reflection is broken.

4.3. Application scenarios

Three classes of scenarios prove particularly productive in school and college settings. The first invites the participant to look inside the process, for instance, to follow a CNC machine through a complete cycle from blank to finished part, with the AR overlay highlighting the moments at which the operator’s decisions actually matter. The second class allows the participant to try a simple assembly operation themselves, performing the work under AR guidance and receiving immediate visual feedback before moving on to the next attempt. The third class is concerned with navigating the working environment as such: where the tools are located, where the safety zone runs, how parts flow through the workstation, and which auxiliary actions occupy more time than the operations they serve. Together these three formats reproduce, in compressed form, the principal modes in which a worker actually inhabits a workplace, and they do so without requiring the participant to commit in advance to any particular profession.

5.1. Technical solution

The platform on which the proposed model is deployed is intentionally mobile and unceremonious. A consumer smartphone, optionally complemented by an inexpensive pair of AR glasses, is sufficient for all the scenarios described above; the only firm technical requirement is cross-platform support, so that a single scenario can be launched on devices running either Android or iOS without modification. A student arriving at a career-guidance event uses their own phone under a bring-your-own-device arrangement, which removes the principal procurement obstacle that has historically prevented schools from experimenting with immersive technologies. Where the context requires reliable identification of the participant — for example, in regional programmes that aggregate results across many events — lightweight decentralised verification schemes based on blockchain and computer vision can be layered onto the same mobile platform without disturbing its bring-your-own-device character

The economics of this configuration are worth stating explicitly. The marginal cost of adding one more participant to a session is essentially zero, since the participant brings their own device. The marginal cost of adding one more profession to the catalogue is the cost of recording and lightly editing a short reference video, which can be done by a partner enterprise with no specialist software. As a result, the total cost of a programme grows almost linearly with the number of distinct professions covered, and almost not at all with the number of students served, which is the inverse of the cost structure that has traditionally constrained career-guidance work.

5.2. Interaction process

The interaction itself unfolds in a manner that participants describe as surprisingly natural. The student stands in front of a training stand or a working area, raises the smartphone or puts on the glasses, and on the screen a step-by-step instruction is superimposed on the real scene in front of them. The eyes do not jump between paper and part: cue and tool sit at the same optical distance, and the gesture suggested by the demonstration is the gesture the participant immediately attempts. Purdue University has confirmed in controlled comparisons that this learning-by-doing format yields levels of comprehension many times higher than those obtained by simply watching the same video on a separate screen. The reason is structural rather than aesthetic: when the demonstration and the action share a common spatial context, the participant does not need to perform the additional cognitive operation of mapping one onto the other, and the cognitive resources thus liberated become available for noticing, for hesitating, and for forming a personal judgement. This structural advantage has been recovered independently in a systematic review of VR and AR applications in vocational education, which associates contextual overlay with the most consistent gains in both affect and performance

5.3. Results

Numbers reported by independent cases such as AVM Technology and SIBUR Digital converge on a consistent picture: motivation among participants rises by approximately forty to fifty per cent compared with conventional career-guidance formats, and the eventual choice of specialty becomes noticeably more deliberate, with fewer participants reporting subsequent regret about their decision. A recent systematic literature review of AR, VR, and intelligent tutoring systems in education and training arrives at a similar aggregate conclusion, highlighting gains in motivation, engagement, and learning outcomes across a wide range of domains

It would be premature to treat these figures as definitive. The samples involved are still modest in size, the methodologies of measurement vary across projects, and the long-term effect on actual occupational trajectories will only become visible as the first cohorts of participants enter the labour market. Nonetheless, the convergence of independent results across very different organisational contexts is encouraging, and it suggests that the observed gains reflect a genuine property of the format rather than the enthusiasm of any individual research group.

The technologies of factory-floor augmented reality can in principle be transferred to the practice of career guidance, but they cannot be transferred directly. The goal of the system has to be reformulated: not the flawless execution of an operation but the conscious, embodied acquaintance of a young person with a profession. Once this reformulation is accepted, the well-documented benefits of industrial AR — the relief of working memory, the contraction of training time, the systematic reduction of errors — remain available, but they begin to serve a different end. They no longer optimise productivity; they enable a richer and more honest form of self-encounter.

The core of the proposed model is contextual guidance, in which the reference and the participant’s own action coincide in time and space. A passive tour-and-observe visit to an enterprise is replaced by a hands-on trial in which the participant temporarily inhabits the role they are considering. What emerges is a small but genuine fragment of work experience, sufficient to anchor a personal judgement that no amount of secondary description can produce.

The principle of minimal authoring is what makes this approach economically realistic at scale. Scenarios are launched on consumer smartphones, three-dimensional development is dispensed with, and the production of new content can be entrusted to partner enterprises rather than to specialised studios. Scaling such a programme to the level of a region or a country becomes a question of organisation, of agreement among schools and enterprises, and of methodological coordination, rather than a question of capital investment. This is a significant practical consideration, since most previous attempts to bring immersive technologies into mass education have collapsed precisely at the point where capital costs met institutional budgets.

The bottom line can be stated briefly. Augmented reality in career guidance is valuable not for the visual effect it produces on the screen but for the action it compels in the world. It places the young person in the position of someone who is doing the work rather than someone who is being told about it, and action of this kind is what reliably triggers the process of self-determination. The visuals are a means; the encounter with one’s own response to the work is the end.

The natural next step in this line of work is the development of adaptive scenarios that adjust their difficulty on the fly in response to the participant’s observed performance, coupled with machine-learning algorithms for the automatic analysis of typical errors. Such an extension would close the loop between the participant’s experience and the long-term refinement of the scenario library, allowing the system to become not only a tool for individual self-discovery but also a continuously improving instrument for understanding how young people actually engage with the world of work.