Community-Integrated Project-Based Learning for Interdisciplinary Engineering Education: A Mechatronics Case Study of a Rideable 5-Inch Gauge Railway

Abstract

This study presents a case of community-integrated project-based learning (PBL) at a Japanese National Institute of Technology (KOSEN). Three students collaborated to design and build a rideable 5-inch gauge railway system, integrating mechanical design, brushless motor control, and computer vision. The project was showcased at public events and a partner high school, providing authentic feedback and enhancing learning relevance. Over 15 weeks, students engaged in hands-on prototyping, interdisciplinary teamwork, and real-world problem-solving. The course design was grounded in four educational frameworks: experiential learning, situated learning, constructive alignment, and self-regulated learning (SRL). SRL refers to students’ ability to plan, monitor, and reflect on their learning—a key skill for managing complex engineering tasks. A mixed-methods evaluation—including surveys, reflections, classroom observations, and communication logs—revealed significant gains in technical competence, engagement, and learner autonomy. Although limited by a small sample size, the study offers detailed insights into how small-scale, resource-conscious PBL can support meaningful interdisciplinary learning and community involvement. This case illustrates how the KOSEN approach, combining technical education with real-world application, can foster both domain-specific and transferable skills, and provides a model for broader implementation of authentic, student-driven engineering education.

1. Introduction

Engineering education increasingly adopts project-based learning (PBL) to foster practical and interdisciplinary skills essential for real-world practice (Krajcik & Shin, 2014; Halverson & Sheridan, 2014). Reviews have highlighted PBL’s effectiveness in promoting active learning, teamwork, and the application of theory to real problems (Lavado-Anguera et al., 2024; Mota et al., 2025). These qualities are vital for future engineers navigating rapid technological and societal change (Prince & Felder, 2006; Brundiers & Wiek, 2013).

However, many PBL implementations remain confined to classroom simulations, limiting authenticity and student motivation (Hmelo-Silver, 2004). Disciplinary silos—such as separating design, electronics, and programming—can hinder holistic learning (Mills & Treagust, 2003). There is increasing recognition that engineering education must also address social relevance by engaging with real communities and users (Brundiers et al., 2010; OECD, 2018).

In response, new PBL models emphasize community engagement, interdisciplinary collaboration, and real-world impact (Bolick et al., 2024; Kondo et al., 2023). These are particularly suited to Japan’s National Institute of Technology (KOSEN) (National Institute of Technology, 2024), which emphasizes hands-on training and local collaboration (Tsutsumi et al., 2024).

This study examines a community-integrated PBL case at a Japanese KOSEN, where students designed a rideable 5-inch gauge railway system. By integrating mechanical design, motor control, and computer vision, the project provided a platform for deep interdisciplinary learning. The findings offer insight into how small-scale, resource-conscious PBL can support meaningful, transferable learning aligned with community needs.

2. Theoretical Background

2.1. Overview

This study draws on four interconnected theoretical frameworks to analyze the educational impact of the railway system project: experiential learning, situated learning, self-regulated learning (SRL), and constructive alignment. Each framework helps explain how students learn through authentic tasks, collaboration, reflection, and alignment. These models were chosen for both their individual and complementary value in supporting PBL at KOSEN. Figure 1 summarizes their integration and their correspondence to the project environment.

2.2. Design-Based and Experiential Learning Foundations

Kolb’s experiential learning theory (Kolb, 1984) views knowledge as developing through cycles of experience, reflection, abstraction, and experimentation. In this project, students repeated construction, discussion, refinement, and testing, supporting both conceptual and technical development.

The iterative design process also exemplifies a design-based learning (DBL) approach, where students engage in continuous problem-solving around real-world constraints (Gómez Puente et al., 2013; Krajcik & Shin, 2014). In this study, students used OpenCV (version 4.9.0) and Python (version 3.11) to build a control system, thus integrating digital tools with mechanical design. These activities match educational efforts to use digital fabrication like 3D printing in STEM curricula (Stern et al., 2019). Such tools increased student ownership, deepened learning, and linked theory to practice.

Overall, these frameworks guided course design by promoting iterative, applied learning through technology-driven design challenges.

2.3. Situated and Scaffolded Learning

The authentic setting—designing a system for community use—illustrates situated learning, where knowledge is built through social participation (Lave & Wenger, 1991). Presenting at public events and collaborating with stakeholders let students practice engineering in a real community.

Vygotsky (1981)’s Zone of Proximal Development (ZPD) highlights the value of scaffolding and peer collaboration. In this project, students supported each other based on their expertise, creating a collaborative environment that encouraged both individual and group progress. Such alignment with real-world audiences and shared tasks increased students’ motivation, the relevance of their work, and the transferability of their skills.

2.4. Self-Regulated Learning

Self-regulated learning (SRL) involves planning, goal-setting, self-monitoring, and adaptation. Zimmerman (2002) identifies SRL as the foundation of autonomous learning. In a PBL context, SRL enables students to handle complex tasks and take ownership of their learning paths (English & Kitsantas, 2013). In this project, students assumed independent roles, communicated via Microsoft TEAMS, and iteratively improved their subsystems. These behaviors demonstrate metacognitive development and ongoing engagement with learning tasks. The explicit use of rubrics and feedback further supported SRL, increasing clarity and self-efficacy (Panadero et al., 2023; Andrade, 2005). This is consistent with other STEAM-based PBL projects, where SRL has been shown to boost motivation and autonomy (López Carrillo et al., 2024). By structuring the course to foster agency, reflection, and iterative problem-solving, the learning design closely followed SRL principles.

2.5. Constructive Alignment

Biggs’s constructive alignment model (Biggs, 1996, 2014) stresses the importance of coherence among learning objectives, teaching methods, and assessment. In this project, targeted learning outcomes—such as integrating mechanical design, electronics, and vision systems—were embedded in course activities and assessments. Student presentations, peer feedback, and system testing offered authentic opportunities to measure these outcomes. The course design avoided isolated tasks, instead offering challenges that required coordinated, interdisciplinary effort. As shown in

Table 1. Mapping of Learning Theories to Project-Based Activities.

Learning Theory	Core Concept	Manifestation in Project
Experiential Learning	Learning through concrete-reflective cycles	Design, test, reflect, iterate cycle with hands-on building
Situated Learning	Authentic context and community participation	Public exhibition, designing for real high school use
Self-Regulated Learning	Goal-setting, monitoring, self-reflection	Independent task planning, timeline control, peer feedback
Constructive Alignment	Coherence between objectives, teaching, assessment	Integration of mechanics, coding, and vision system; peer review
Zone of Proximal Development (ZPD)	Learning through expert scaffolding	Peer mentoring, task-sharing based on expertise gaps
Design-Based Learning (DBL)	Iterative design for real-world problems	Real-time control system using OpenCV and motor integration

Taken together, these theories offer a robust framework for understanding how students learn by doing, collaborating, and reflecting within an aligned, authentic, and technology-rich environment.

, this alignment ensured that learning processes, instructional supports, and evaluation mechanisms all worked together toward the same educational goals.

3. Context and Project Overview

3.1. Educational Setting

This study was conducted in the “Advanced Precision Design Engineering” course at a Japanese National Institute of Technology (KOSEN). The course aimed to foster interdisciplinary design thinking and applied problem-solving skills. A 15-week project-based learning (PBL) format was adopted (see

Table 2. 15-Week Curriculum Schedule.

Week	Lecture Topics	Project Activities
1	Precision Measurement Introduction	Project Kick-off and Team Formation
2	Length and Angle Measurement	Initial Concept Sketches
3	Error and Tolerance	Subsystem Planning
4	Sensor Calibration	Frame Design (SolidWorks)
5	Signal Processing Basics	Drive System Design
6	Data Logging Methods	Circuit Layout
7	Analog-to-Digital Conversion	Motor Selection and Control Setup
8	Feedback and Control Theory	Simulation and Assembly Planning
9	OpenCV Basics	Camera Setup
10	Image Filtering and Edge Detection	Curve Detection Algorithm Design
11	Pattern Recognition	Python and Arduino Integration
12	Robustness Testing	Subsystem Integration
13	User Testing Preparation	Pilot Operation and Feedback Collection
14	System Improvement	Final Debugging
15	Presentation Preparation	Final Presentation and Peer Review

), combining lecture-based instruction with team-based design and prototyping (Biggs, 1996; Gómez Puente et al., 2013). Each weekly session (90 minutes) included both theoretical instruction and hands-on implementation. The curriculum progressed from basic measurement and CAD modeling to motor control and image processing using Python and OpenCV. This integration reflects the design-based learning model (Krajcik & Shin, 2014) and aligns with constructive alignment theory (Biggs, 2014). The inclusion of public presentations and community outreach further reinforced real-world relevance and experiential learning (Brundiers & Wiek, 2013; Halverson & Sheridan, 2014). Similar STEAM-oriented practices had previously been implemented at the same institution, such as a bamboo helicopter project integrating 3D CAD and 3D printing (Tsutsumi et al., 2024), providing a foundation for this more advanced PBL project.

3.2. Project Objectives

The project tasked students with designing and constructing a rideable 5-inch gauge railway system capable of pulling two passenger coaches and carrying approximately ten individuals. The system was shown at local events and a partner high school, providing an authentic audience for feedback and testing (Brundiers et al., 2010). The design goals included a target speed of 5 km/h (walking pace), a brushless DC motor with at least 3 Nm of torque, continuous operation for up to 10 hours, and modular maintainability. Importantly, the system needed to be safe, reliable, and transportable for public demonstrations. These constraints offered a concrete engineering challenge rooted in real-world application, consistent with previous implementations of community-based PBL (Ríos et al., 2010).

3.3. Technical Overview

The railway system was divided into four key technical subsystems:

• Mechanical design: A steel chassis and wheel assembly modeled in SolidWorks (2024) CAD software (see Figure 2)
• Motor control: A brushless DC motor (3 Nm) connected via a 1:13 gear ratio and controlled with an off-the-shelf driver
• Power unit: A modular lead-acid battery array with swappable units
• Vision system: A real-time curve detection algorithm implemented in Python using OpenCV (see Figure 3 and Figure 4)

Integrating design, control, and sensing under real constraints offers a holistic approach, seldom seen in past studies (cf. Kondo et al., 2023; Stern et al., 2019). Students emphasized reproducibility and modular design for long-term educational value.

3.4. Collaborative Roles

Three students participated in the course, each assuming a distinct technical role based on individual expertise:

• Mechanical Lead: Responsible for system architecture and CAD modeling (SolidWorks)
• Circuit Lead: Managed wiring, motor control integration, and procurement
• Vision Lead: Developed image processing and system-level integration in Python

This role-based collaboration promoted accountability, distributed expertise, and peer teaching, in line with frameworks for cooperative learning in engineering education (Gómez Puente et al., 2013). Weekly meetings and milestone presentations facilitated feedback and helped synchronize efforts. These interactions illustrate the value of real-world application and community feedback in enhancing socially relevant, situated learning experiences (Brundiers et al., 2010).

4. Methodology

4.1. Research Design

This study employed a mixed-methods research design to examine the learning outcomes of a small-scale, community-integrated PBL course. The primary data sources included a 5-item post-project Likert-scale questionnaire (Cronbach’s alpha = 0.72), open-ended student reflections, instructor observations, and Microsoft TEAMS communication logs. These sources allowed triangulation of evidence from cognitive, technical, and behavioral domains. The questionnaire was adapted from validated PBL instruments (Kokotsaki et al., 2016) to assess student collaboration, technical development, and self-regulated learning. Qualitative data provided contextual insights into student experiences, peer dynamics, and project challenges. Given the small sample size (n = 3), the study adopts an exploratory case study approach aimed at generating detailed, context-rich descriptions rather than statistical generalization. This methodology prioritizes depth over breadth and is suited for capturing the nuanced learning processes and social interactions within a resource-conscious, interdisciplinary engineering project.

4.2. Participants and Course Implementation

The participants were three students enrolled in the Advanced Course in Mechanical and Information Systems Engineering at a Japanese National Institute of Technology (KOSEN). Each student had prior experience with CAD, programming, and electronics from earlier coursework. Informed by research on technology-supported learning design (Ifenthaler & Yau, 2020), the course utilized a hybrid model that balanced flexibility with structured collaboration. Team roles were assigned based on student interests and competencies, fostering ownership and balanced collaboration. This approach is consistent with self-regulated learning principles and cooperative learning models in engineering education.

4.3. Data Collection Instruments

To assess the learning process and outcomes, four main data sources were used:

• Post-course questionnaire: A 5-item Likert-scale survey, supplemented by two open-ended reflection prompts.
• Instructor observations: Field notes recorded during prototyping sessions and final presentations.
• Student artifacts: CAD models, Python code, and subsystem integration designs.
• Communication logs: Archived records capturing discussions, planning, and problem-solving behavior.

These instruments allowed data collection from multiple perspectives and helped identify patterns in student engagement, collaboration, and technical development.

Table 3. Overview of Data Collection Instruments and Analysis Methods.

Data Source	Collection Method	Type	Analysis Method
Questionnaire	End-of-course form	Quantitative	Descriptive statistics
Open-ended responses	Written by students	Qualitative	Thematic analysis
Observations	Instructor notes	Qualitative	Analytical memoing
TEAMS chat logs	Archived communication	Qualitative	Discourse analysis

provides an overview of the data sources, collection methods, data types, and analytical approaches used in the study.

4.4. Data Analysis

Quantitative data from the questionnaire were analyzed using descriptive statistics to identify trends in student engagement and skill development. For qualitative data, thematic analysis was applied to student reflections and instructor observations, focusing on recurring themes such as technical growth, collaboration, and real-world application. Communication logs from Microsoft TEAMS were examined through discourse analysis to trace patterns of planning, initiative, and self-regulation. This triangulated approach aligns with established PBL evaluation practices (Kokotsaki et al., 2016; Mills & Treagust, 2003). The analytical framework drew on prior studies of design-based and experiential learning (Gómez Puente et al., 2013; Krajcik & Shin, 2014) and was guided by theories of self-regulated learning and collaborative reflection (Zimmerman, 2002; English & Kitsantas, 2013). The use of TEAMS also reflects broader trends in digital support for hybrid PBL environments (Zhong & Lyu, 2022). While the small sample size limits generalization, the case offers rich insights into how community-oriented, technology-enhanced PBL can be implemented in vocational education.

5. Results and Analysis

5.1. Quantitative Survey Results

All three students completed a 5-item Likert-scale questionnaire at the end of the project. Item-by-item analysis indicated consistently high scores for engagement and skill development (Appendix A.2). These quantitative findings were substantiated by qualitative data, such as students’ reflections and communication logs. Scores ranged from 4.3 to 5.0 out of 5.0 on a five-point Likert scale, indicating high engagement and perceived growth. Students gave high ratings for collaboration, technical skill development, and motivation, and felt confident applying their learning to real-world situations. These results were not analyzed for statistical significance but served as entry points for deeper qualitative insights. For example, top-rated items like “I collaborated effectively” and “I developed new technical skills” aligned with student reflections and chat records. One noted, “We divided our roles naturally but helped each other when stuck.” Another wrote, “I had never used a motor driver before, but through trial and error I got it to work.” These comments reflect meaningful learning through peer support, productive struggle, and hands-on problem-solving—hallmarks of effective project-based environments (Panadero et al., 2023).

5.2. Qualitative Reflections

Three high school teachers who attended the final demonstration praised the project’s educational value. One noted, “Students showed strong interest, and it would be beneficial in other STEM classes.” Another remarked, “Interacting with real mechanical systems was exciting and inspiring.” These comments highlight the project’s broader educational impact and support the value of socially embedded learning (Brundiers et al., 2010; Brundiers & Wiek, 2013). Student reflections echoed this, with one stating, “Tuning the vision system was difficult but rewarding. I learned through trial and error.” Overcoming challenges through hands-on experimentation reflects Kolb (1984)’s experiential learning cycle, especially concrete experience and active experimentation. Another commented, “Managing the wiring made me feel like a real engineer.” This comment illustrates situated learning, where authentic tasks in a community help develop identity and make learning more relevant.

Such narratives emphasize authentic, situated learning (Lave & Wenger, 1991) and highlight the motivational effects of engaging with real-world engineering challenges (see Figure 5).

5.3. TEAMS Communication Analysis

Archived communication logs (with selected excerpts presented in Appendix D) provided further evidence of student initiative and planning. For example, messages between students and the instructor demonstrated proactive management of procurement and scheduling:

Student A: “Some parts are missing. I’ll prepare a list.”

Teacher: “Understood. Let me know when it’s ready.”

Such exchanges demonstrate professional-level communication, responsibility, and self-regulation—key components of engineering competence and workplace readiness (Gómez Puente et al., 2013; Zimmerman, 2002).

While not addressing TEAMS specifically, recent learning analytics research emphasizes how digital platforms support asynchronous coordination and engagement (Ifenthaler & Yau, 2020).

5.4. Observational Evidence (Photographs)

Photographic documentation confirmed students’ active participation (see Figure 6). Images captured moments of prototyping, peer-to-peer instruction, and system testing (see Figure 6). These visuals complement the communication logs and survey responses, offering tangible evidence of collaborative, problem-driven engagement.

Key takeaways include:

• Students actively assembling and testing mechanical components
• Clear evidence of peer-to-peer instruction and task coordination
• Final system deployment in an authentic community setting

5.5. Theoretical Alignment

The findings closely align with the theoretical framework outlined in Section 2. Kolb (1984)’s experiential learning was reflected in the iterative design-build-test cycle. Constructive alignment was evident in the coherence between goals, instruction, and assessment (Biggs, 1996). Situated learning emerged through authentic, community-based activities (Lave & Wenger, 1991). Self-regulated learning was seen in students’ ownership of tasks and timeline management (Zimmerman, 2002). The use of digital tools further supported learner agency (Ifenthaler & Yau, 2020). Together, these elements demonstrate that the learning environment effectively supported technical, cognitive, and social development as intended.

6. Discussion

6.1. Educational Implications and Theoretical Contributions

This study demonstrates that integrating project-based learning (PBL) with community engagement fosters both technical skills and broader competencies such as collaboration, communication, and self-regulated learning. These outcomes align with key STEM education goals, including hands-on design, creativity, and problem-solving (Halverson & Sheridan, 2014; Sawyer, 2006). Presenting the railway vehicle to a real audience enhanced authenticity and student motivation (Robinson, 2011; Brundiers & Wiek, 2013). Although the project involved a small team, it produced rich evidence of meaningful learning, consistent with prior findings (Hidayat et al., 2024). However, the unique institutional and cultural context limits its generalizability (Ramírez De Dampierre et al., 2024). Student reflections and observations align closely with the theoretical framework described in Section 2. Their engagement illustrates Kolb (1984)’s experiential learning cycle, Vygotsky’s concept of scaffolding, and the integrative nature of design-based learning (Tsai et al., 2022; Krajcik & Shin, 2014). These findings support the educational value of authentic, interdisciplinary, and socially embedded learning environments. Future implementations may benefit from structured time management tools, scaffolding for heterogeneous learners, and advanced planning for logistics and community collaboration to improve scalability and equity.

6.2. Limitations and Future Directions

Despite its contributions, this study has several limitations. First, the small sample size (n = 3) restricts generalizability. However, as an exploratory case study, it aimed to provide in-depth insights into learning processes and team dynamics. Second, the setting—a Japanese KOSEN—offers a valuable example of hands-on engineering education, but further studies are needed to assess its applicability across other institutional and cultural contexts. Third, data collection relied primarily on self-reports, observations, and communication records. Incorporating performance-based assessments, peer evaluations, or follow-up interviews would yield a more comprehensive picture of learning outcomes. Future research should consider larger samples, longitudinal designs, and the role of educational technologies in supporting self-regulated learning and collaboration. Despite these limitations, the study provides a valuable case of how community-connected, interdisciplinary PBL can foster deep learning in resource-constrained environments.

These challenges underscore the need for careful planning and institutional support when adapting such models to larger student populations.Bias from the author’s dual role was mitigated through data triangulation.

7. Conclusions

This study shows that community-integrated PBL can offer valuable interdisciplinary learning, even with limited resources and a small student team. Students developed technical, collaborative, and self-regulation skills while engaging with real-world challenges and community needs. Despite the limited scale, the results suggest that small-scale PBL can foster deep learning and student motivation when linked to authentic contexts.