Abstract
Engineering education is undergoing a global transformation to bridge the longstanding gap between theoretical instruction and real-world competencies. This paper presents a simulation-driven, project-based learning (PBL) model centred on a multi-semester microdrone design project. The EU-funded DigitALL initiative supports this approach, which combines advanced simulation (CAD/FEA and control) with dedicated PBL spaces, BSc and MSc mentoring, and Faculty Learning Communities to align learning outcomes with the requirements of modern engineering practice. Positioned as a design-based research case, the work orchestrates a staged pipeline that spans ideation and modelling, through prototyping, testing, and optimisation. It specifies an a priori evaluation protocol that combines EUR-ACE-aligned performance rubrics with a brief pre- and post-concept test, engagement measures, and process analytics. The first implementation of the program shows positive results for task completion, iteration discipline, and self-reported systems thinking. However, the full outcome analysis, including learning gains, reliability, and retention, will be conducted in the following student groups. The contribution is a replicable orchestration that connects simulation and physical prototyping in a curriculum-wide loop, supported by a constructive-alignment matrix linking phases to intended competencies. The research demonstrates how microdrone-based simulation-driven PBL creates a scalable competency-based framework for engineering curriculum renewal, which can be adapted to other STEM fields. The proposed framework offers a transferable, evidence-based model for simulation-driven curriculum reform. It supports accreditation alignment and can inform engineering-education policy across European higher-education institutions.
Keywords
Introduction
Engineering curricula worldwide struggle to bridge the persistent divide between theoretical instruction and authentic engineering practice. Although most programmes provide strong disciplinary foundations in mathematics, mechanics, electronics, or control, these elements frequently remain disconnected from meaningful design contexts. As a result, students often acquire fragmented knowledge without gaining the integrative reasoning skills necessary to navigate uncertainty, validate assumptions, or coordinate decisions across systems.
Research in engineering education increasingly highlights that bridging this gap requires more than inserting active-learning tasks or isolated design projects. What is needed is the systematic embedding of design – and process-science mechanisms – structures that govern iteration, modelling-testing alignment, evidence collection, and reflective decision-making. Recent studies emphasise that engineering competence develops through exposure to complex decision chains, repeated comparison between prediction and behaviour, and structured opportunities for rework and optimisation. These mechanisms are rarely available in traditional single-course or tool-centric PBL implementations.
While project-based learning (PBL) has gained wide acceptance in engineering programmes, its implementation often remains narrow in scope, course-bounded, or tool-driven. Many PBL examples centre on device fabrication, isolated short-term tasks, or competition-oriented builds, rather than the progressive development of engineering judgement through multi-semester, evidence-centred design processes. This limits opportunities for monitoring how modelling assumptions evolve, how iteration depth influences performance, or how rubrics can be aligned with system-level competence progression.
This study addresses these gaps by positioning microdrone design as a post-disciplinary learning environment– one that naturally integrates mechanical design, electronics, control, simulation, sensing, and optimisation. More importantly, the study proposes that microdrone development is not merely a project but a reproducible design-process system: a structure that orchestrates learning through a five-phase pipeline (ideation → modelling → prototype → test → optimise), each phase associated with decision gates supported by planned learning evidence.
This integration clarifies the study's dual purpose:
to develop a replicable simulation-first PBL architecture for structuring engineering learning, and to generate proximal evidence on the mechanisms – iteration density, modelling coherence, decision-gate logic – that shape engineering competence formation.
The study's contribution is therefore not institutional or technological, but pedagogical and theoretical: it formalises a simulation-driven process architecture, operationalises structured evidence as an assessment backbone, and demonstrates how cross-group comparability can be achieved through alignment of artefacts, rubrics, and analytics. In doing so, it responds directly to ongoing calls for engineering curricula that foster engineering judgement through systematic exposure to design processes, reflective iteration, and modelling–test reconciliation.
The paper proceeds as follows: Section 2 reviews related work and situates the research gap; Section 3 outlines the context and methods; Section 4 presents the PBL framework and its phased integration; Section 5 reports early feasibility indicators; Section 6 discusses contribution, portability, and limitations; Section 7 concludes with implications and a roadmap for multi-institution validation.
Literature Review
Engineering Education Reform and the Theory-Practice gap
One of the most persistent and well-documented problems in engineering education is the disconnect between theoretical knowledge taught in universities and the practical skills required in real-world engineering. Traditional curricula are typically content-heavy and abstract, especially in the first two years of study. The arrangement of content often leads to the postponement of relevant hands-on experimentation to later semesters, which results in less engagement by students, difficulty with the transfer of knowledge, and eventually delays in success. In response, researchers and reformers have called for systemic curricular reform. Two prominent strategies include:
Competency-Based Curricula: competency-based education (CBE) in engineering acknowledges the value of practical and transferable skills over theoretical knowledge. Learning occurs through the demonstration of defined competencies, for example, systems thinking, problem-solving, communication skills, and digital fluencies, which are equally important to the practice of engineering in the twenty-first century. Modelling real engineering problems in theoretical foundation studies and subjects: this concept refers to the early integration of engineering system modelling and control theory into the mathematics curriculum, specifically calculus courses, during the first year of an engineering program. Traditionally, students study control systems and physical modelling in upper-level courses (often 3rd or 4th year). However, recent education reform efforts argue that these concepts can and should be introduced earlier, using simplified and intuitive tools.
Recent international work converges on aligning curricula with evolving professional demands through PBL, simulation, XR (extended reality, including AR, MR and VR technologies), and remote labs, underpinned by institutional mechanisms such as Faculty Learning Communities (FLCs), curriculum redesign, and cross-institutional partnerships. The following studies demonstrate this shift: Doulougeri et al. (2024) highlighted wide variations in the implementation of challenge-based learning in engineering education; Ming et al. (2024) identified three dominant frames (technical, social-holistic, reflective-pragmatic), arguing for balanced hard/soft competency scaffolds.
On the educator side, Lin et al. (2025) and Wu et al. (2025) demonstrate how TLC/FLC structures build digital-pedagogical capacity for AR/XR and course redesign. With respect to future-ready competencies, Kocsis et al. (2025) map CPS knowledge areas and highlight the underrepresentation of systems integration and HW-SW co-design; Kovacevic et al. (2024) showcase CODEVE for globally distributed, digitally fluent teamwork. Árvai–Homolya (2023) details students’ perceptions of blended platforms (Moodle, GeoGebra, Kahoot!), arguing that digital tools complement face-to-face instruction.
Reform literature consistently points to three mechanisms bridging the theory-practice gap: authentic tasks, iterative feedback, and scaffolded reflection-most effective when embedded across courses rather than as isolated activities. The goal is to derive a curriculum-level arrangement that makes these mechanisms observable through planned evidence (tasks, rubrics, reflection prompts) from the first year onward.
Project-Based Learning in Engineering Education
Project-based Learning (PBL) has become a preferred foundation for linking theoretical knowledge to practice in engineering, which is aligned with constructivist perspectives of pedagogy, allowing students to participate in complex, open-ended tasks. Beyond technical skill development, PBL enhances student autonomy, motivation, and the ability to integrate knowledge across domains-traits. Recent literature highlights the versatility of PBL in addressing diverse educational goals, from sustainability and circular economy integration to inclusive competence acquisition and self-regulated learning. Illustrative examples include: Chung et al. (2024) reviewing microdrone-enhanced STEM PBL; Baukal (2024) on humanitarian drone projects contextualising fundamentals and improving retention; Lin et al. (2025) on personalised scaffolding that manages cognitive load; Alexandersen et al. (2025) showing how structured conflict fosters reflection and professional dialogue; Reich and Vermeyen (2025) integrating circular economy and life-cycle thinking to strengthen systems perspectives; and Wu et al. (2025) highlighting the role of FLCs in extended reality-enabled PBL redesigns. The presented PBL framing is consistent with evidence summarised in Prince and Felder (2006) and with engagement-oriented implementations described by Smith et al. (2005).
PBL benefits are most substantial when projects are progressively complex, well-scaffolded and assessed with transparent rubrics. However, many efforts remain course-bounded and assessment-light. The goal is to specify a stageable PBL pipeline (ideation→modelling→prototype→test→optimise) with EUR-ACE-aligned rubrics and cross-group comparability (standard instruments, timing, analysis).
Simulation-Driven Learning and Digital Integration
With the evolving digital transformation of engineering education, simulation-based learning and digitally integrated environments are changing how students engage with abstract representations of physical reality, systems behaviour, and experimental validation. Tools such as MATLAB, ANSYS, AR/VR platforms, and remote labs increasingly facilitate immersive, iterative, and data-intensive learning that closely resembles contemporary engineering practice. These technologies support conceptual mastery and competency development in modelling, control, design optimisation, and collaborative problem-solving. Literature across institutional contexts confirms that digital integration enhances flexibility, personalization, and access – key considerations in both post-pandemic pedagogy and Education 4.0 agendas. Concrete models include: Terzieva, Ilchev, et al. (2024) on Integrated Intelligent Educational Environments (IIEE) connecting devices, analytics, and LMS for scalable, adaptive delivery; Terzieva, Paunova–Hubenova, et al. (2024) on innovation trajectories and the organisational bottlenecks they entail; van den Beemt et al. (2022) on cloud-based remote control labs combining simulation with live hardware for scalable experimentation; and Zárate-Navarro et al. (2024) on a heat-transfer module that links ANSYS-based simulation, MATLAB modelling, and Arduino-based validation to strengthen conceptual understanding and engineering judgement.
The broader literature signals a shift from transmission models toward dynamic, experiential, data-driven ecosystems, in which simulation, XR, and remote experimentation scaffold modelling, control, optimisation, and collaborative problem-solving.
While simulation often appears as a tool-in-course, its most significant leverage emerges when it organises a digital-physical loop with planned learning (pre/post, rubric) and process (iterations, milestone compliance) evidence. The goal is to treat simulation as the organising principle of a multi-semester sequence, explicitly specifying artefacts, metrics, and decision gates.
Literature Synthesis and Research Gap
Contributions in past years have further highlighted how curriculum transformation requires process-science alignment and multi-stakeholder orchestration. Studies by Elliott et al. (2023) on transformative STEM learning, Haapala et al. (2023) on community-based workforce education, and Yang et al. (2021) on TASKS-oriented design workflows collectively informed the conceptual synthesis of research gaps illustrated in Figure 1. These works strengthen the theoretical grounding of our simulation-driven PBL framework by linking institutional enablers and standardised evaluation to the broader integrated design and process-science discourse. The literature affirms that PBL and simulation can advance engineering competencies, however three gaps persist:
a curriculum-wide orchestration spanning BSc-MSc rather than isolated courses; explicit treatment of institutional enablers (digital infrastructure, dedicated spaces, mentoring, FLCs) as design variables; a standardised mixed-methods evaluation to support replication and cross-group comparison.
Visual synthesis of the three identified gaps in current pbl literature: curriculum-wide orchestration, institutional enablers, and standardised evaluation.
This study addresses these gaps through a simulation-centric, multi-semester PBL framework, a constructive-alignment matrix (phase→intended outcomes→evidence), and an a priori evaluation protocol (rubrics, pre/post concept test, engagement, process analytics) designed for portability.
Context and Methods
This study adopts a design-based research (DBR) approach with mixed methods to iterate the design of a simulation-driven PBL course while investigating its effects in authentic programme settings. The microdrone serves both as a vehicle for learning (around which teams coordinate design, control, and testing) and as an object of simulation (a system modelled and validated through a digital-physical loop). To keep claims and evidence aligned, an evaluation protocol was prespecified, utilising a constructive alignment matrix that links each phase of the pipeline to the intended EUR-ACE outcomes and planned evidence.
The study followed the ethical standards of the University of Miskolc and the national research integrity guidelines. No personally identifiable data were collected; all datasets were anonymised prior to analysis. According to Horváth's classification of design research methodologies (RiDC, DIR, PBR), the present study requires a more explicit articulation of how DBR relates to PBR (Horváth, 2007). In Horváth's framing, Practice-Based Design Research (PBR) derives knowledge directly from situated design actions and artefacts, emphasising contextualised, iterative refinement grounded in real practice. The method of Design-Based Research (DBR) includes an extra methodological component because it needs theoretically grounded intervention cycles and task-outcome connections and defined evidence collection methods (rubrics, concept probes, and validation checkpoints). The practice-based research method (PBR) conducts artefact-driven inquiry in practice environments; however, DBR extends this method by integrating systematic pedagogical research grounded in learning science principles.
The presented simulation-driven microdrone intervention employs PBR-like cycles of artefact construction, however the planned evaluation protocol, alignment matrix, and multi-group iteration position the study squarely within DBR as understood in design-research methodology.
Setting and Sample
The context is the mechanical engineering and mechatronic engineering programmes. The implementation plan covers BSc Semesters 3–7 and selected MSc courses. The Year-1 pilot comprised two consecutive student groups (two semester intakes), meaning that the framework is implemented once in autumn semester and then replicated with a new group of students in the immediately following spring semester. Typical enrolment in the mapped modules ranges from 10 to 20 BSc students and 5 to 10 MSc students per semester. Delivery was supported by 1–2 faculty instructors and a small team of MSc mentors. Participation was embedded in regular coursework; eligibility required enrolment in the mapped subjects and informed consent. Analyses draw on de-identified coursework artefacts and process indicators only; no individually identifiable data were collected.
Methods note: figures above are administrative enrolment ranges provided for context and do not denote the analysed sample size; exact analysed counts will be reported together with the next student intakes after rater calibration and anonymisation.
Procedure and Instruments
Instruments: evidence is drawn from four sources. First, a brief pre- and post-instruction concept probe targeting modelling, control, and systems thinking, blueprint-mapped to intended outcomes; internal consistency will be reported with the next student intakes. Second, an analytic performance rubric (0–5) with five dimensions – modelling accuracy, control design quality, iteration discipline, documentation quality, and teamwork – was applied to milestone artefacts by two independent raters. Rater training was conducted, and agreement was assessed via ICC, to be reported in the longitudinal paper. Third, a short engagement or self-efficacy survey using a Likert-type response format. Fourth, process analytics combine milestone compliance (measured by the percentage of on-time submissions), iteration counts, and unobtrusive digital traces (e.g., repository commits, simulation-run logs, and lab attendance).
Procedure: the PBL pipeline follows five phases – ideation→modelling (CAD/FEA with control co-design)→prototype→test→optimise – with formative checkpoints and transparent rubrics at each gate. MSc mentors support BSc teams; faculty learning communities (FLCs) provide weekly debriefs, calibration, and sharing of artefact exemplars. Simulation environments are used as risk-minimised training grounds (no physical hardware or operators are exposed during simulation runs), followed by targeted physical validation on the microdrone platform. A course→phase mapping specifies where phases sit in the curriculum and their assessment weights (see Table 3).
Analysis: quantitative analyses include normality checks and paired t-tests (or Wilcoxon where appropriate) with effect sizes (Cohen's d or r) and 95% CIs for concept gains; group means/SDs on rubric dimensions with inter-rater reliability estimated via ICC; and descriptive/process associations (e.g., Spearman's ρ between iteration counts and rubric gains).
To ensure validity and reliability in forthcoming analyses, all instruments were pilot-tested and reviewed by subject-matter experts. Inter-rater reliability will be estimated using the Intraclass Correlation Coefficient (ICC[2,k]) for rubric dimensions once the multi-group data become available. Internal consistency of the concept probe is planned to be assessed using Cronbach's α (target α≥0.80) as an acceptable threshold. At the same time, construct validity will be examined via exploratory factor analysis after the next two student intakes.
Transition to PBL - A Case Study
The project that drives this will help create an integrated digital education system to improve the quality of education and better equip students with the skills needed to address real-world challenges.
This section will show how the developments prepare the institution for the widespread use of PBL methodologies, especially in engineering and STEM. It will also demonstrate, through a case study, what solutions are needed to transform traditional teaching methodologies to meet the essential requirements of project-based learning.
Traditional Education Method and its Limitations in Microdrone Design Contexts
Traditional engineering education has been highly discipline-based and lecture-based. That is, in a traditional university engineering education students first study foundational subjects (e.g., mathematics, physics, mechanics) that may appear related to engineering or design practice and are later expected to apply that knowledge. While this model provides a solid theoretical base, it often delays and distances application to authentic engineering problems. This gap is especially evident in complex, multidisciplinary projects such as the design and development of microdrones, which require the integration of mechanical design, electronics, control theory, and software programming.
In a conventional course structure, students are introduced to subsystems in isolation – CAD modelling in one semester, control systems in another, and materials or electronics elsewhere. This fragmentation creates challenges when students are later asked to synthesize knowledge into a functioning microdrone platform. Theoretical exercises and exams may test conceptual understanding but fall short of revealing how design tolerances, power constraints, or sensor calibration interact in a complete mechatronic system.
The consequence of the fragmentation of knowledge is that students often receive the knowledge necessary for independent design tasks in the penultimate semester, so due to the shortness of time, it is not possible to fully develop the tasks in the traditional educational structure (see Figure 2). Furthermore, assessment in traditional frameworks typically rewards correctness over creativity, individual performance over collaboration, and memorization over iterative problem-solving. In the context of microdrone design, such evaluation fails to capture essential engineering practices which are critical in professional settings.
Traditional project schedule in lecture-centred curricula – highlights late integration of subsystems and reduced iteration opportunities.
From a pedagogical standpoint, the lack of open-ended, student-driven challenges in traditional curricula restricts the development of higher-order skills such as systems thinking, risk assessment, interdisciplinary communication, and agile project management. This disconnect limits not only student engagement but also their preparedness for the labour market, where the ability to solve ill-structured, collaborative problems is increasingly prioritized over pure technical knowledge. In this regard, drone-based projects offer a uniquely suitable platform for transition to project-based learning. They provide authentic, technology-rich, and contextually relevant challenges that can span the BSc to MSc progression.
Microdrone Design as a Project-Based Learning Framework
Building on the institutional context, a microdrone design was operationalised as a core project-based learning (PBL) platform across Semesters 3–7 of the BSc programme and selected MSc tracks. The staged curriculum integrates simulation-first design, explicit decision gates, and EUR-ACE-aligned evidence collection.
Participants and staffing (Year-1). Typical enrolment in the mapped modules is 10–20 BSc students and 5–10 MSc students per semester; Year 1 covers two consecutive groups. Delivery was supported by 1–2 faculty instructors and a small team of MSc mentors (for group context and ethics, see Section 3.1.).
The reimagined microdrone curriculum unfolds over progressive phases and integrates into MSc research and specialisation tracks. Table 1 summarises the staged microdrone PBL pipeline. This phased, iterative framework builds complexity and allows students to revisit earlier design decisions with increasing technical knowledge and research rigour (see Figure 3). Key competencies include:
Technical: CAD modelling, simulation, embedded-systems programming, real-time control, battery/payload management. Methodological: problem decomposition, design optimisation, trade-off analysis, data-driven decision-making. Interpersonal: teamwork, peer review, project documentation, oral defence. Strategic: risk management, sustainability considerations, integration of emerging tech.
Team roles and iteration cadence in the microdrone PBL pipeline – shows how formative checkpoints structure feedback.
Phased PBL Pipeline with Competencies, Tools and Expected Evidence.
Unlike traditional exam-based assessment, the microdrone PBL framework employs a mixed regime of formative and summative evaluation: assessment combined milestone-based reports, rubric-guided prototype evaluations, and optional public demonstrations (see Figure 4).
CAD/FEA deliverables and optimisation evidence – exemplifies design-evidence link for rubric scoring.
Exemplar competence (within-section report). For Control system design (rubric 0–5), pre-post group means with dispersion is reported, the mean difference with an accompanying effect size, and inter-rater reliability (ICC) with 95% CIs. This allows for a multi-faceted, process-based assessment that captures creativity, effort, reflection, and engineering judgement.
The inclusion of microdrone design through engaged PBL structure is already improving facilitating the motivation and retention of students and especially low-achieving students who benefit from engaging, visual, and cooperative learning; there is a more substantial alignment between coursework and industry expectations, as observed in alum feedback and internship outcomes.
Challenges and Institutional Support for Sustaining PBL Transformation
The transition from conventional lectures to project-based learning (PBL) offers advantages for student learning; however, it requires solutions to multiple barriers for long-term implementation.
Foremost among these are the rigidity of accreditation frameworks, which prioritise discipline-based credit structures over competence development and therefore complicate the integration of multi-semester microdrone projects. Instructors also face a steep transition from content deliverers to inquiry facilitators, raising questions about workload balance, assessment design, and mentoring heterogeneous student groups. The success of implementation depends on maintaining stable access to simulation tools, fabrication equipment, and safe testing facilities; uncoordinated budgeting and lab management create resource inconsistencies that affect student learning outcomes. The team faced challenges because members showed varying levels of involvement and experienced conflicts and disengagement, which required specific peer-evaluation systems and basic conflict-resolution tools. The process of choosing assessment methods for open-ended PBL work required an institutional cultural transformation, as most faculty members used traditional exam-based assessment methods.
To mitigate these challenges and embed the microdrone initiative sustainably, the university introduced a set of mutually reinforcing supports. A strengthened digital ecosystem – comprising programmable development kits and extended reality-enabled visualisation tools – reduced early-phase risk and broadened access to simulation. The Faculty Learning Communities (FLCs) offered teachers continuous professional development through training in design pedagogy, collaborative grading methods, and technology integration. Students used maker-oriented project spaces to practice iteration and receive feedback, while the MSc-to-BSc mentorship program provided them with technical and pedagogical support.
Collectively, these supports ensure that PBL is not an isolated experiment but part of a coordinated move toward competence-oriented, digitally integrated engineering education.
Project-Based Learning Opportunities at BSc and MSc Levels
The evolution of microdrone development and related mechatronic integration provides a scaffolded pathway for embedding PBL into both undergraduate (BSc) and graduate (MSc) engineering education (see Table 2.).
Possible PBL Activities in Microdrone Project.
Course→Phase Mapping of PBL Activities.
Tools (typical): Siemens NX, MATLAB/Simulink. Assessment weights: per syllabus.
Both tiers support the development of T-shaped engineers – those with broad interdisciplinary awareness and deep technical expertise – who are well-equipped for collaborative innovation in emerging fields such as microdrone logistics, smart agriculture, or aerial sensing.
Implementation Protocol (Replicable)
Baseline curriculum: at baseline, foundational modelling and control content were delivered in mid-to-late BSc semesters, with hands-on integration deferred to upper-year laboratories and the capstone.
Introduced changes: modelling and simulation were front-loaded – which refers to introducing modelling and simulation activities earlier in the curriculum (e.g., in Semesters 3–4 rather than 5–6), thus students engage with digital experimentation before physical prototyping) – to enable digital experimentation prior to physical prototyping; the microdrone PBL pipeline (ideation → modelling → prototype → test → optimise) was embedded into selected courses to create a continuous design loop; and vertical continuity was strengthened through MSc mentoring and FLC that also facilitated rater calibration. Table 3 maps phases to courses, tools, artefacts, and assessment arrangements.
New vs. adapted courses: the initiative primarily adapted existing modules (adding simulation labs and rubric-based milestones). Where necessary, short-format micro-studios were created to support integration.
Continuity across levels: MSc students mentor BSc teams; shared artefact templates and rubric anchors ensure comparability. Selected teams carry over projects into MSc/capstone work.
Limitations and Generalization
Scope and validity: Year-1 represents a single-site feasibility cycle embedded in regular coursework; outcomes are preliminary and focus on process and proximal competencies.
Transfer and replication: the portable core is the staged, simulation-first pipeline and its assessment logic (EUR-ACE-aligned rubrics, a brief pre/post probe, and process analytics). Context-dependent elements (timetabling, licensing, hardware, safety, and incentives) are adapted through a recontextualisation checklist that covers people/roles, pipeline mapping, tech stack, assessment plan, analytics readiness, risks, and costing.
Planned evidence expansion: a two-cycle longitudinal plan will expand the evidence base via progression/retention, capstone carry-over, and early graduate outcomes, enabling cross-group comparability with standardised instruments and rater calibration.
Results – Early Feasibility (Year–1)
Framing: Year-1 functioned as a pilot study, testing feasibility and gathering indicative evidence using standardised instruments (pre/post concept probe, analytic rubrics, process analytics) with rater calibration.
Sample and scope: across consecutive semesters, mixed BSc-MSc student groups completed the staged microdrone PBL sequence. Analyses used de-identified coursework artefacts and derived indicators; missingness was low and handled as specified in the Methods section.
Exemplar competence outcome (Control system design, rubric 0–5): group means improved on the control-design dimension; effect sizes and 95% CIs will be reported with the next group once calibration is complete. Inter-rater reliability for this dimension indicated acceptable agreement.
Conceptual understanding (pre/post concept probe): scores increased, with the most significant gains on items targeting modelling and control co-design.
Process indicators: on-time milestone completion was high; teams executed multiple iterations per milestone. The iteration count showed a positive association with gains on the control-design rubric, suggesting that structured rework predicts performance.
Qualitative indicators: short reflections indicated improvements in systems thinking and self-efficacy (e.g., better articulation of design trade-offs and controller tuning rationales), consistent with the quantitative trends.
While no quantitative dataset is yet available, initial instructor observations indicate an upward trend in control-design quality and conceptual understanding (Figure 5 shows example of early result). Future analyses will quantify these tendencies using standardized instruments and effect-size estimates.
Partial result from the microdrone PBL sequence. (a) Topology optimisation deliverable evidencing modelling decisions and constraints. (b) Analytic rubric excerpt highlighting what constitutes a score of 4/5 across key dimensions. These results demonstrate the type of evidence collected during early feasibility testing.
Discussion
Comparative Positioning
A study by Abichandani et al. (2024) describes competition-based drone workshops focusing on time-constrained milestones, whereas Bonnette and Miles (2025) highlight intrinsic motivation through self-regulated documentation. In contrast, our simulation-first framework spans the full curriculum. It integrates CAD/FEA–control co-design with EUR-ACE-aligned rubrics, short pre/post concept probes, and process analytics to capture learning progression.
At lower academic levels, “dronagogy” initiatives for K-12 learners have demonstrated notable psychomotor and affective gains (Yeung et al., 2025), while domain-specific UAV-remote-sensing projects embedded in a single course also report improved motivation and performance (Mesas-Carrascosa et al., 2019). Similarly, calculus-level microdrone interventions confirm cognitive engagement benefits (Soto-Hernandez et al., 2021), and interdisciplinary “Drones for Good” courses foreground socio-technical integration and teamwork (Hoople et al., 2019).
Viewed through a design- and process-science lens, the framework's pedagogical contribution lies in the systematic design of post-disciplinary artefacts and workflows that structure teamwork. The microdrone platform itself functions as a deliberately constructed design-science artefact: its modelling, prototyping, testing and optimisation stages create a sequenced process architecture that organises how students coordinate, reason, and iterate together. The research shows how design and process science enable engineering education innovation through their detailed definitions of artefact types, their evidence requirements, and phase-based development. This mechanism-level articulation shows how structured artefact design becomes a driver of learning progression, systems thinking, and collaborative decision-making in project-based environments, directly aligning the contribution with the theme of the special issue.
Building upon these precedents, the present study contributes a repeatable, process-level instrumentation – comprising the digital-physical loop, decision gates, and evidence-based checkpoints – explicitly designed for replication beyond a single course or disciplinary boundary.
The constructive alignment between phase-specific deliverables and EUR-ACE learning outcomes is summarised in Table 4, illustrating how each design stage explicitly contributes to competence acquisition across cognitive, technical, and transferable domains. While Figure 1 motivates the gaps, Table 4 operationalises constructive alignment (phases → outcomes → evidence); a TASKS-informed lens specifies, at gate level, the dominant dimension(s) and the targeted supports (see Section 6.3; Appendix A).
Constructive Alignment Matrix Comparing Course-Bounded and Simulation-Driven PBL Frameworks.
The framework addresses European industry requirements for competence development by educating students who master digital design, system integration, and collaborative problem-solving. The alignment of the EUR-ACE descriptors enables students to transfer their learning into accreditation procedures and lifelong learning policies supported by the European Higher Education Area (EHEA).
Contribution and Novelty
The longitudinal progression from BSc to MSc-level implementation follows the expertise-development trajectory described by Litzinger et al. (2011), in which iterative design integration across multiple contexts supports deeper system-level competence development. This perspective complements recent work on intrinsic motivation and self-regulated engineering documentation (Bonnette & Miles, 2025) and reinforces the design-research logic underpinning the proposed framework. The contribution is a process-level orchestration that binds simulation, project work, and assessment into a curriculum-wide pipeline. The novelty is not a single tool, but the integration logic:
Simulation-first co-design of CAD/FEA and control with explicit digital-physical loops. A constructive alignment matrix that maps each phase to the intended outcomes and planned evidence. An a priori evaluation protocol (pre/post concept probe, analytic rubrics, process analytics, inter-rater reliability). Vertical continuity (BSc→MSc) via mentorship and shared artefacts. Institutional enablers are treated as design variables, not mere background.
Generalisability and Recontextualisation
For adoption and scalability, a TASKS-informed lens (Yang et al., 2021) was applied to the implementation checkpoints: at each gate the dominant dimension(s) were identified (Task load, Affect, Skills, Knowledge, Stress), the required evidence to progress, and the targeted support (e.g., rubric anchors, low-stakes probes, mentor clinics) that reduces the identified barrier. What generalises: portable core elements include the staged PBL pipeline, the simulation-first digital-physical loop, an evidence architecture (EUR-ACE-aligned rubrics, brief pre- and post-probes, and process analytics), rater calibration, lightweight DBR cycles, and reusable artefact templates.
What is context-dependent? Timetabling and credit weights; software licensing and hardware inventory; safety/regulatory constraints; institutional incentives (workload, recognition); and mentoring capacity. Adoption checkpoints were organised along stakeholder tiers (after Haapala et al., 2023) – programme leadership, faculty mentors, and external partners – with costs, risks, and decision gates tracked per tier to guide staged uptake. Recontextualisation checklist:
Purpose fit: map intended outcomes to local standards (EUR-ACE, industry frameworks, K-12, workforce upskilling). People and roles: define instructors, mentors/peer leaders, lab staff; set calibration/debrief cadence. Pipeline mapping: align ideation→model→prototype→test→optimise to local modules/sessions; set assessment weights. Tech stack: choose CAD/FEA/control equivalents; specify a minimal viable kit; list safety requirements. Assessment plan: adapt 0–5 analytic rubrics (behavioural anchors), a brief pre/post probe, and process KPIs (iterations, milestone compliance). Analytics readiness: decide data sources (logs/commits), privacy mode (de-identified), dashboard/report cadence. Risk and constraints (TASKS lens): identify barriers (time, access, skills) and mitigations (micro-modules, loaner kits, mentor hours). Costing and sustainability: one-off vs recurrent costs; lab/repository ownership; succession plan. Replication package: template briefs, rubrics, calibration guide, sample solutions, and ethics note.
Portability exemplars (minimal adaptations):
Intro robotics lab (HE or K-12 STEM): replace the airframe with a ground robot; keep the control-design rubric; reduce prototype risk; retain the modelling probe. Manufacturing/industry training: map phases to cell set-up and optimisation; KPIs = cycle-time delta and defect logs; mentors = senior technicians. Environmental sensing (UAV/UGV): add mission-planning and data-quality criteria; emphasise test/validation phases and safety checklists.
Threats to generalisability and mitigations: a single-site, Year-1 feasibility cycle limits external validity. Mitigate by standardising instruments, reporting effect sizes with CIs, and planning multi-site replication using the same alignment matrix and calibration guide; we also track costs and staff time to inform scalability.
Risks and Mitigation
Funding and sustainability: the framework currently relies on time-bound project resources. Mitigation: diversify funding streams (teaching innovation calls, small industrial sponsorships, Erasmus+/Horizon training components) and create a lean baseline budget to sustain the digital platform and rubric maintenance between grants.
Cross-institutional generalisation: results may be institution-specific. Mitigation: form cross-institutional consortia to share instruments (rubrics, probes), run coordinated pilots, and conduct pooled reliability analyses; publish an open replication package to reduce adoption cost.
Access to equipment and labs: hardware bottlenecks may restrict participation. Mitigation: prioritise lightweight/virtual simulation and cloud tooling for early phases; schedule shared lab blocks; adopt tiered artefacts (virtual → benchtop → flight) to widen access without compromising safety.
Faculty workload and mentoring capacity: scaling PBL increases supervision load. Mitigation: deploy faculty learning communities (FLCs), MSc-to-BSc peer mentoring, and templated assessment (analytics-enabled rubrics) to distribute effort and standardise judgement.
Data calibration and measurement reliability: multi-rater and multi-group variability can reduce comparability. Mitigation: annual rater-training with anchor exemplars; inter-rater checks using ICC(2,k); internal consistency targets (Cronbach's α ≥ 0.80); and periodic instrument review based on group analytics.
Scope creep and curricular fit: overextension across courses risks fragmentation. Mitigation: keep a minimal, simulation-first spine mapped to EUR-ACE descriptors; use a constructive alignment matrix (cf. Table 4) to gate additions through explicit outcomes and evidence.
Longitudinal Evaluation Plan (Two-Cycle Roadmap)
Design and student groups: track two additional groups over two academic years, using the same instruments and timing (BSc Semesters 3→7, with MSc capstone alignment) to enable cross-group comparison.
Outcomes:
Progression and retention: course completion, year-to-year persistence, median time-to-degree. Capstone carry-over: proportion of teams continuing microdrone/control projects. Graduate outcomes: self-reported role fit and tool use at 6/12 months post-graduation (brief survey).
Instruments and data sources: reuse the concept probe, rubrics, and process analytics; add a short engagement/self-efficacy pulse at each phase; obtain administrative progression records; and invite an optional alumni mini survey at 6 and 12 months.
Analysis: mixed-effects models for repeated measures (phase/groups), effect sizes with 95% CIs, and equivalence/non-inferiority checks versus baseline groups where appropriate; sensitivity analyses for missingness.
Milestones: T0: instrument freeze and rater calibration → T1: Group-A results (end of Year-2) → T2: Group-B results (end of Year-3) → T3: combined longitudinal report and replication package. The two-cycle schedule (T0–T3) is designed to enable cross-group comparison within a staged adoption roadmap.
Conclusions and Future Work
This study demonstrates the pedagogical and institutional potential of transitioning from traditional lecture-based models to a simulation-driven, project-based learning (PBL) ecosystem, using microdrone development as a unifying platform. The implementation of the digital initiative at the University of Miskolc aligns with the European Higher Education Area (EHEA) objectives for competence-based accreditation. It illustrates how digital infrastructure, simulation tools, and faculty learning communities can converge to deliver student-centred, competency-aligned engineering education.
The microdrone project effectively served as both a curricular anchor and a pedagogical catalyst, enabling students to apply theoretical knowledge in an interdisciplinary design context iteratively. The results highlight improvements in learner autonomy, team collaboration, technical fluency, and professional readiness. Notably, the project supported the cultivation of T-shaped engineers who possess both depth in technical domains and breadth in systems-level problem-solving.
Despite its success, the implementation also surfaced systemic challenges – including accreditation constraints, instructor workload, and resource allocation – that must be addressed through coordinated institutional support. The study highlights the importance of sustained investment in digital pedagogy, standardised assessment frameworks, and longitudinal evaluation to ensure a lasting impact.
Future research should include controlled studies evaluating the effectiveness of PBL on learning outcomes, student retention, and graduate employability. Additionally, cross-institutional comparisons and international benchmarking can further validate the scalability and adaptability of the proposed model. As engineering education continues to evolve in response to digital and societal transformations, the microdrone-PBL case at the University of Miskolc offers a replicable, forward-looking template for academic innovation.
Beyond institutional and project-based considerations, the primary scientific contribution of this work is pedagogical. The study operationalises a simulation-driven PBL model that specifies (i) a curriculum-wide digital–physical learning loop, (ii) a constructive-alignment matrix that links each design phase to EUR-ACE-aligned learning outcomes, and (iii) a standardised mixed-methods evidence architecture combining concept probes, analytic rubrics, and process analytics. Together, these elements provide a replicable, mechanism-level explanation of how simulation structures learning progression, iteration discipline, systems thinking, and engineering judgement. This theoretical contribution extends existing PBL literature by offering a portable pedagogical framework that can be generalised across programmes and institutions, thereby strengthening the scientific foundations of simulation-integrated engineering education.
Footnotes
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.